diff options
author | A. Unique TensorFlower <gardener@tensorflow.org> | 2016-11-16 13:39:57 -0800 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2016-11-16 13:46:36 -0800 |
commit | 93411a010af67c4d6c233245dd0acc4d43e01cdd (patch) | |
tree | 713533b1c2e10952af893cfc9eaa113e7186615b /tensorflow/g3doc | |
parent | 54afc49e09b4b24b166cbad8c115f794c149c06f (diff) |
Update generated Python Op docs.
Change: 139372651
Diffstat (limited to 'tensorflow/g3doc')
205 files changed, 1667 insertions, 1561 deletions
diff --git a/tensorflow/g3doc/api_docs/python/array_ops.md b/tensorflow/g3doc/api_docs/python/array_ops.md index d03441a636..8949cf9468 100644 --- a/tensorflow/g3doc/api_docs/python/array_ops.md +++ b/tensorflow/g3doc/api_docs/python/array_ops.md @@ -24,14 +24,14 @@ results in a rounded value.) ##### Args: -* <b>`string_tensor`</b>: A `Tensor` of type `string`. +* <b>`string_tensor`</b>: An `Output` of type `string`. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.float32, tf.int32`. Defaults to `tf.float32`. The numeric type to interpret each string in `string_tensor` as. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `out_type`. + A `Output` of type `out_type`. A Tensor of the same shape as the input `string_tensor`. @@ -200,13 +200,13 @@ endian orderings will give different results. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. * <b>`type`</b>: A `tf.DType` from: `tf.float32, tf.float64, tf.int64, tf.int32, tf.uint8, tf.uint16, tf.int16, tf.int8, tf.complex64, tf.complex128, tf.qint8, tf.quint8, tf.qint32, tf.half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `type`. + A `Output` of type `type`. - - - @@ -276,13 +276,13 @@ This operation returns N 1-D integer tensors representing shape of `input[i]s`. ##### Args: -* <b>`input`</b>: A list of at least 1 `Tensor` objects of the same type. +* <b>`input`</b>: A list of at least 1 `Output` objects of the same type. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A list with the same number of `Tensor` objects as `input` of `Tensor` objects of type out_type. + A list with the same number of `Output` objects as `input` of `Output` objects of type out_type. - - - @@ -415,14 +415,14 @@ reshape(t, []) ==> 7 ##### Args: -* <b>`tensor`</b>: A `Tensor`. -* <b>`shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`tensor`</b>: A `Output`. +* <b>`shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Defines the shape of the output tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. + A `Output`. Has the same type as `tensor`. - - - @@ -512,15 +512,15 @@ size 1. ##### Args: -* <b>`input`</b>: A `Tensor`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 0-D (scalar). Specifies the axisension index at which to expand the shape of `input`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Contains the same data as `input`, but its shape has an additional axisension of size 1 added. @@ -832,14 +832,14 @@ dimension. For example, tiling `[a b c d]` by `[2]` produces ##### Args: -* <b>`input`</b>: A `Tensor`. 1-D or higher. -* <b>`multiples`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. 1-D or higher. +* <b>`multiples`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. Length must be the same as the number of dimensions in `input` * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -1213,8 +1213,8 @@ output[2:, :, 3, :, ...] = input[2:, :, 3, :, ...] ##### Args: -* <b>`input`</b>: A `Tensor`. The input to reverse. -* <b>`seq_lengths`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. The input to reverse. +* <b>`seq_lengths`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with length `input.dims(batch_dim)` and `max(seq_lengths) < input.dims(seq_dim)` * <b>`seq_dim`</b>: An `int`. The dimension which is partially reversed. @@ -1224,7 +1224,7 @@ output[2:, :, 3, :, ...] = input[2:, :, 3, :, ...] ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The partially reversed input. It has the same shape as `input`. @@ -1282,14 +1282,14 @@ reverse(t, dims) ==> [[[[8, 9, 10, 11], ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. Up to 8-D. -* <b>`dims`</b>: A `Tensor` of type `bool`. 1-D. The dimensions to reverse. +* <b>`dims`</b>: An `Output` of type `bool`. 1-D. The dimensions to reverse. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. The same shape as `tensor`. + A `Output`. Has the same type as `tensor`. The same shape as `tensor`. - - - @@ -1345,15 +1345,15 @@ reverse(t, dims) ==> [[[[8, 9, 10, 11], ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. Up to 8-D. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. The indices of the dimensions to reverse. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. The same shape as `tensor`. + A `Output`. Has the same type as `tensor`. The same shape as `tensor`. - - - @@ -1417,7 +1417,7 @@ Extract `patches` from `images` and put them in the "depth" output dimension. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 4-D Tensor with shape `[batch, in_rows, in_cols, depth]`. * <b>`ksizes`</b>: A list of `ints` that has length `>= 4`. The size of the sliding window for each dimension of `images`. @@ -1445,7 +1445,7 @@ Extract `patches` from `images` and put them in the "depth" output dimension. ##### Returns: - A `Tensor`. Has the same type as `images`. + A `Output`. Has the same type as `images`. 4-D Tensor with shape `[batch, out_rows, out_cols, ksize_rows * ksize_cols * depth]` containing image patches with size `ksize_rows x ksize_cols x depth` vectorized in the "depth" dimension. @@ -1469,12 +1469,12 @@ precise description. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`, where spatial_shape has `M` dimensions. -* <b>`block_shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`block_shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with shape `[M]`, all values must be >= 1. -* <b>`paddings`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`paddings`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D with shape `[M, 2]`, all values must be >= 0. `paddings[i] = [pad_start, pad_end]` specifies the padding for input dimension `i + 1`, which corresponds to spatial dimension `i`. It is required that @@ -1588,7 +1588,7 @@ precise description. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -1608,8 +1608,8 @@ block size. ##### Args: -* <b>`input`</b>: A `Tensor`. 4-D with shape `[batch, height, width, depth]`. -* <b>`paddings`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. 4-D with shape `[batch, height, width, depth]`. +* <b>`paddings`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies the padding of the input with zeros across the spatial dimensions as follows: @@ -1701,7 +1701,7 @@ block size. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -1758,12 +1758,12 @@ reverse of SpaceToBatch. See below for a precise description. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`, where spatial_shape has M dimensions. -* <b>`block_shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`block_shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with shape `[M]`, all values must be >= 1. -* <b>`crops`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`crops`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D with shape `[M, 2]`, all values must be >= 0. `crops[i] = [crop_start, crop_end]` specifies the amount to crop from input dimension `i + 1`, which corresponds to spatial dimension `i`. It is @@ -1878,7 +1878,7 @@ reverse of SpaceToBatch. See below for a precise description. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -1898,11 +1898,11 @@ followed by cropping along the `height` and `width` dimensions. ##### Args: -* <b>`input`</b>: A `Tensor`. 4-D tensor with shape +* <b>`input`</b>: A `Output`. 4-D tensor with shape `[batch*block_size*block_size, height_pad/block_size, width_pad/block_size, depth]`. Note that the batch size of the input tensor must be divisible by `block_size * block_size`. -* <b>`crops`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`crops`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies how many elements to crop from the intermediate result across the spatial dimensions as follows: @@ -1914,7 +1914,7 @@ followed by cropping along the `height` and `width` dimensions. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. 4-D with shape `[batch, height, width, depth]`, where: height = height_pad - crop_top - crop_bottom @@ -2065,13 +2065,13 @@ x = [[[[1, 2, 3, 4], ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`block_size`</b>: An `int` that is `>= 2`. The size of the spatial block. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -2162,14 +2162,14 @@ x = [[ [1], [2], [5], [6]], ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`block_size`</b>: An `int` that is `>= 2`. The size of the spatial block, same as in Space2Depth. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -2202,14 +2202,14 @@ this operation will permute `params` accordingly. ##### Args: -* <b>`params`</b>: A `Tensor`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`params`</b>: A `Output`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. * <b>`validate_indices`</b>: An optional `bool`. Defaults to `True`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `params`. + A `Output`. Has the same type as `params`. - - - @@ -2313,14 +2313,14 @@ Batched indexing into a 3-tensor: ##### Args: -* <b>`params`</b>: A `Tensor`. `P-D`. The tensor from which to gather values. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`params`</b>: A `Output`. `P-D`. The tensor from which to gather values. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. `Q-D`. Index tensor having shape `[d_0, ..., d_{Q-2}, K]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `params`. + A `Output`. Has the same type as `params`. `(P+Q-K-1)-D`. Values from `params` gathered from indices given by `indices`. @@ -2352,17 +2352,17 @@ count ==> [2, 1, 3, 1, 2] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. +* <b>`x`</b>: A `Output`. 1-D. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (y, idx, count). + A tuple of `Output` objects (y, idx, count). -* <b>`y`</b>: A `Tensor`. Has the same type as `x`. 1-D. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. -* <b>`count`</b>: A `Tensor` of type `out_idx`. 1-D. +* <b>`y`</b>: A `Output`. Has the same type as `x`. 1-D. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. +* <b>`count`</b>: A `Output` of type `out_idx`. 1-D. - - - @@ -2378,7 +2378,7 @@ operator which extracts values or slices from a given tensor. TODO(simister): Add a link to Variable.__getitem__ documentation on slice syntax. -`shape` is a `TensorShape` with rank `P` and `indices` is a `Tensor` of rank +`shape` is a `TensorShape` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `shape`. @@ -2445,19 +2445,19 @@ The resulting tensor would look like this: ##### Args: -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. +* <b>`updates`</b>: A `Output`. A Tensor. Must have the same type as tensor. A tensor of updated values to store in ref. -* <b>`shape`</b>: A `Tensor`. Must have the same type as `indices`. +* <b>`shape`</b>: A `Output`. Must have the same type as `indices`. A vector. The shape of the resulting tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `updates`. + A `Output`. Has the same type as `updates`. A new tensor with the given shape and updates applied according to the indices. @@ -2507,8 +2507,8 @@ For example: ##### Args: -* <b>`data`</b>: A `Tensor`. -* <b>`partitions`</b>: A `Tensor` of type `int32`. +* <b>`data`</b>: A `Output`. +* <b>`partitions`</b>: An `Output` of type `int32`. Any shape. Indices in the range `[0, num_partitions)`. * <b>`num_partitions`</b>: An `int` that is `>= 1`. The number of partitions to output. @@ -2516,7 +2516,7 @@ For example: ##### Returns: - A list of `num_partitions` `Tensor` objects of the same type as data. + A list of `num_partitions` `Output` objects of the same type as data. - - - @@ -2572,13 +2572,13 @@ For example: ##### Args: -* <b>`indices`</b>: A list of at least 1 `Tensor` objects of type `int32`. -* <b>`data`</b>: A list with the same number of `Tensor` objects as `indices` of `Tensor` objects of the same type. +* <b>`indices`</b>: A list of at least 1 `Output` objects of type `int32`. +* <b>`data`</b>: A list with the same number of `Output` objects as `indices` of `Output` objects of the same type. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. - - - @@ -2837,17 +2837,17 @@ result = range_min + ((input - numeric_limits<T>::min()) * range_scale) ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. -* <b>`min_range`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`min_range`</b>: An `Output` of type `float32`. The minimum scalar value possibly produced for the input. -* <b>`max_range`</b>: A `Tensor` of type `float32`. +* <b>`max_range`</b>: An `Output` of type `float32`. The maximum scalar value possibly produced for the input. * <b>`mode`</b>: An optional `string` from: `"MIN_COMBINED", "MIN_FIRST"`. Defaults to `"MIN_COMBINED"`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. - - - @@ -2909,10 +2909,10 @@ operations that have to perform further calculations on them. ##### Args: -* <b>`input`</b>: A `Tensor` of type `float32`. -* <b>`min_range`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: An `Output` of type `float32`. +* <b>`min_range`</b>: An `Output` of type `float32`. The minimum scalar value possibly produced for the input. -* <b>`max_range`</b>: A `Tensor` of type `float32`. +* <b>`max_range`</b>: An `Output` of type `float32`. The maximum scalar value possibly produced for the input. * <b>`T`</b>: A `tf.DType` from: `tf.qint8, tf.quint8, tf.qint16, tf.quint16, tf.qint32`. * <b>`mode`</b>: An optional `string` from: `"MIN_COMBINED", "MIN_FIRST"`. Defaults to `"MIN_COMBINED"`. @@ -2920,11 +2920,11 @@ operations that have to perform further calculations on them. ##### Returns: - A tuple of `Tensor` objects (output, output_min, output_max). + A tuple of `Output` objects (output, output_min, output_max). -* <b>`output`</b>: A `Tensor` of type `T`. The quantized data produced from the float input. -* <b>`output_min`</b>: A `Tensor` of type `float32`. The actual minimum scalar value used for the output. -* <b>`output_max`</b>: A `Tensor` of type `float32`. The actual maximum scalar value used for the output. +* <b>`output`</b>: A `Output` of type `T`. The quantized data produced from the float input. +* <b>`output_min`</b>: An `Output` of type `float32`. The actual minimum scalar value used for the output. +* <b>`output_max`</b>: An `Output` of type `float32`. The actual maximum scalar value used for the output. - - - @@ -2936,27 +2936,27 @@ Concatenates quantized tensors along one dimension. ##### Args: -* <b>`concat_dim`</b>: A `Tensor` of type `int32`. +* <b>`concat_dim`</b>: An `Output` of type `int32`. 0-D. The dimension along which to concatenate. Must be in the range [0, rank(values)). -* <b>`values`</b>: A list of at least 2 `Tensor` objects of the same type. +* <b>`values`</b>: A list of at least 2 `Output` objects of the same type. The `N` Tensors to concatenate. Their ranks and types must match, and their sizes must match in all dimensions except `concat_dim`. -* <b>`input_mins`</b>: A list with the same number of `Tensor` objects as `values` of `Tensor` objects of type `float32`. +* <b>`input_mins`</b>: A list with the same number of `Output` objects as `values` of `Output` objects of type `float32`. The minimum scalar values for each of the input tensors. -* <b>`input_maxes`</b>: A list with the same number of `Tensor` objects as `values` of `Tensor` objects of type `float32`. +* <b>`input_maxes`</b>: A list with the same number of `Output` objects as `values` of `Output` objects of type `float32`. The maximum scalar values for each of the input tensors. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (output, output_min, output_max). + A tuple of `Output` objects (output, output_min, output_max). -* <b>`output`</b>: A `Tensor`. Has the same type as `values`. A `Tensor` with the concatenation of values stacked along the +* <b>`output`</b>: A `Output`. Has the same type as `values`. An `Output` with the concatenation of values stacked along the `concat_dim` dimension. This tensor's shape matches that of `values` except in `concat_dim` where it has the sum of the sizes. -* <b>`output_min`</b>: A `Tensor` of type `float32`. The float value that the minimum quantized output value represents. -* <b>`output_max`</b>: A `Tensor` of type `float32`. The float value that the maximum quantized output value represents. +* <b>`output_min`</b>: An `Output` of type `float32`. The float value that the minimum quantized output value represents. +* <b>`output_max`</b>: An `Output` of type `float32`. The float value that the maximum quantized output value represents. - - - @@ -2990,17 +2990,17 @@ idx ==> [1, 3, 5] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. Values to keep. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. 1-D. Values to remove. +* <b>`x`</b>: A `Output`. 1-D. Values to keep. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. 1-D. Values to remove. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (out, idx). + A tuple of `Output` objects (out, idx). -* <b>`out`</b>: A `Tensor`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. +* <b>`out`</b>: A `Output`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. @@ -3022,14 +3022,14 @@ Quantization is called fake since the output is still in floating point. ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. * <b>`min`</b>: An optional `float`. Defaults to `-6`. * <b>`max`</b>: An optional `float`. Defaults to `6`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. - - - @@ -3041,9 +3041,9 @@ Compute gradients for a FakeQuantWithMinMaxArgs operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxArgs operation. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxArgs operation. * <b>`min`</b>: An optional `float`. Defaults to `-6`. * <b>`max`</b>: An optional `float`. Defaults to `6`. @@ -3051,7 +3051,7 @@ Compute gradients for a FakeQuantWithMinMaxArgs operation. ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. Backpropagated gradients below the FakeQuantWithMinMaxArgs operation: `gradients * (inputs >= min && inputs <= max)`. @@ -3074,14 +3074,14 @@ This operation has a gradient and thus allows for training `min` and `max` value ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. - - - @@ -3093,24 +3093,24 @@ Compute gradients for a FakeQuantWithMinMaxVars operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxVars operation. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxVars operation. min, max: Quantization interval, scalar floats. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). + A tuple of `Output` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). -* <b>`backprops_wrt_input`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. inputs: +* <b>`backprops_wrt_input`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. inputs: `gradients * (inputs >= min && inputs <= max)`. -* <b>`backprop_wrt_min`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. min parameter: +* <b>`backprop_wrt_min`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. min parameter: `sum(gradients * (inputs < min))`. -* <b>`backprop_wrt_max`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. max parameter: +* <b>`backprop_wrt_max`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. max parameter: `sum(gradients * (inputs > max))`. @@ -3132,14 +3132,14 @@ This operation has a gradient and thus allows for training `min` and `max` value ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. - - - @@ -3151,27 +3151,27 @@ Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxVars operation, shape one of: `[d]`, `[b, d]`, `[b, h, w, d]`. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxVars operation, shape same as `gradients`. min, max: Quantization interval, floats of shape `[d]`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). + A tuple of `Output` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). -* <b>`backprops_wrt_input`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. inputs, shape same as +* <b>`backprops_wrt_input`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. inputs, shape same as `inputs`: `gradients * (inputs >= min && inputs <= max)`. -* <b>`backprop_wrt_min`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. min parameter, shape `[d]`: +* <b>`backprop_wrt_min`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. min parameter, shape `[d]`: `sum_per_d(gradients * (inputs < min))`. -* <b>`backprop_wrt_max`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. max parameter, shape `[d]`: +* <b>`backprop_wrt_max`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. max parameter, shape `[d]`: `sum_per_d(gradients * (inputs > max))`. diff --git a/tensorflow/g3doc/api_docs/python/constant_op.md b/tensorflow/g3doc/api_docs/python/constant_op.md index e39c9961e5..43889ac02d 100644 --- a/tensorflow/g3doc/api_docs/python/constant_op.md +++ b/tensorflow/g3doc/api_docs/python/constant_op.md @@ -152,9 +152,9 @@ fill([2, 3], 9) ==> [[9, 9, 9] ##### Args: -* <b>`dims`</b>: A `Tensor` of type `int32`. +* <b>`dims`</b>: An `Output` of type `int32`. 1-D. Represents the shape of the output tensor. -* <b>`value`</b>: A `Tensor`. 0-D (scalar). Value to fill the returned tensor. +* <b>`value`</b>: A `Output`. 0-D (scalar). Value to fill the returned tensor. @compatibility(numpy) Equivalent to np.full @@ -164,7 +164,7 @@ fill([2, 3], 9) ==> [[9, 9, 9] ##### Returns: - A `Tensor`. Has the same type as `value`. + A `Output`. Has the same type as `value`. @@ -243,17 +243,17 @@ tf.linspace(10.0, 12.0, 3, name="linspace") => [ 10.0 11.0 12.0] ##### Args: -* <b>`start`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`start`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. First entry in the range. -* <b>`stop`</b>: A `Tensor`. Must have the same type as `start`. +* <b>`stop`</b>: A `Output`. Must have the same type as `start`. Last entry in the range. -* <b>`num`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`num`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Number of values to generate. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `start`. 1-D. The generated values. + A `Output`. Has the same type as `start`. 1-D. The generated values. diff --git a/tensorflow/g3doc/api_docs/python/contrib.learn.md b/tensorflow/g3doc/api_docs/python/contrib.learn.md index 6c23a336c8..2da052b54f 100644 --- a/tensorflow/g3doc/api_docs/python/contrib.learn.md +++ b/tensorflow/g3doc/api_docs/python/contrib.learn.md @@ -2043,68 +2043,32 @@ Perform various training, evaluation, and inference actions on a graph. ### `class tf.contrib.learn.RunConfig` {#RunConfig} -This class specifies the specific configurations for the run. +This class specifies the configurations for an `Estimator` run. -If you're a Google-internal user using command line flags with learn_runner.py -(for instance, to do distributed training or to use parameter servers), you -probably want to use learn_runner.EstimatorConfig instead. +If you're a Google-internal user using command line flags with +`learn_runner.py` (for instance, to do distributed training or to use +parameter servers), you probably want to use `learn_runner.EstimatorConfig` +instead. - - - -#### `tf.contrib.learn.RunConfig.__init__(master=None, task=None, num_ps_replicas=None, num_cores=0, log_device_placement=False, gpu_memory_fraction=1, cluster_spec=None, tf_random_seed=None, save_summary_steps=100, save_checkpoints_secs=600, save_checkpoints_steps=None, keep_checkpoint_max=5, keep_checkpoint_every_n_hours=10000, job_name=None, is_chief=None, evaluation_master='')` {#RunConfig.__init__} +#### `tf.contrib.learn.RunConfig.__init__(master=None, num_cores=0, log_device_placement=False, gpu_memory_fraction=1, tf_random_seed=None, save_summary_steps=100, save_checkpoints_secs=600, save_checkpoints_steps=None, keep_checkpoint_max=5, keep_checkpoint_every_n_hours=10000, evaluation_master='')` {#RunConfig.__init__} Constructor. -If set to None, `master`, `task`, `num_ps_replicas`, `cluster_spec`, -`job_name`, and `is_chief` are set based on the TF_CONFIG environment -variable, if the pertinent information is present; otherwise, the defaults -listed in the Args section apply. - -The TF_CONFIG environment variable is a JSON object with two relevant -attributes: `task` and `cluster_spec`. `cluster_spec` is a JSON serialized -version of the Python dict described in server_lib.py. `task` has two -attributes: `type` and `index`, where `type` can be any of the task types -in the cluster_spec. When TF_CONFIG contains said information, the -following properties are set on this class: - - * `job_name` is set to [`task`][`type`] - * `task` is set to [`task`][`index`] - * `cluster_spec` is parsed from [`cluster`] - * 'master' is determined by looking up `job_name` and `task` in the - cluster_spec. - * `num_ps_replicas` is set by counting the number of nodes listed - in the `ps` job of `cluster_spec`. - * `is_chief`: true when `job_name` == "master" and `task` == 0. - -Example: -``` - cluster = {'ps': ['host1:2222', 'host2:2222'], - 'worker': ['host3:2222', 'host4:2222', 'host5:2222']} - os.environ['TF_CONFIG'] = json.dumps({ - {'cluster': cluster, - 'task': {'type': 'worker', 'index': 1}}}) - config = RunConfig() - assert config.master == 'host4:2222' - assert config.task == 1 - assert config.num_ps_replicas == 2 - assert config.cluster_spec == server_lib.ClusterSpec(cluster) - assert config.job_name == 'worker' - assert not config.is_chief -``` +Note that the superclass `ClusterConfig` may set properties like +`cluster_spec`, `is_chief`, `master` (if `None` in the args), +`num_ps_replicas`, `task_id`, and `task_type` based on the `TF_CONFIG` +environment variable. See `ClusterConfig` for more details. ##### Args: * <b>`master`</b>: TensorFlow master. Defaults to empty string for local. -* <b>`task`</b>: Task id of the replica running the training (default: 0). -* <b>`num_ps_replicas`</b>: Number of parameter server tasks to use (default: 0). * <b>`num_cores`</b>: Number of cores to be used. If 0, the system picks an appropriate number (default: 0). * <b>`log_device_placement`</b>: Log the op placement to devices (default: False). * <b>`gpu_memory_fraction`</b>: Fraction of GPU memory used by the process on each GPU uniformly on the same machine. -* <b>`cluster_spec`</b>: a `tf.train.ClusterSpec` object that describes the cluster - in the case of distributed computation. If missing, reasonable - assumptions are made for the addresses of jobs. * <b>`tf_random_seed`</b>: Random seed for TensorFlow initializers. Setting this value allows consistency between reruns. * <b>`save_summary_steps`</b>: Save summaries every this many steps. @@ -2119,17 +2083,36 @@ Example: * <b>`keep_checkpoint_every_n_hours`</b>: Number of hours between each checkpoint to be saved. The default value of 10,000 hours effectively disables the feature. -* <b>`job_name`</b>: the type of task, e.g., 'ps', 'worker', etc. The `job_name` - must exist in the `cluster_spec.jobs`. -* <b>`is_chief`</b>: whether or not this task (as identified by the other parameters) - should be the chief task. * <b>`evaluation_master`</b>: the master on which to perform evaluation. -##### Raises: + +- - - + +#### `tf.contrib.learn.RunConfig.cluster_spec` {#RunConfig.cluster_spec} -* <b>`ValueError`</b>: if num_ps_replicas and cluster_spec are set (cluster_spec - may come from the TF_CONFIG environment variable). + + +- - - + +#### `tf.contrib.learn.RunConfig.evaluation_master` {#RunConfig.evaluation_master} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.get_task_id()` {#RunConfig.get_task_id} + +Returns task index from `TF_CONFIG` environmental variable. + +If you have a ClusterConfig instance, you can just access its task_id +property instead of calling this function and re-parsing the environmental +variable. + +##### Returns: + + `TF_CONFIG['task']['index']`. Defaults to 0. - - - @@ -2141,7 +2124,28 @@ Example: - - - -#### `tf.contrib.learn.RunConfig.job_name` {#RunConfig.job_name} +#### `tf.contrib.learn.RunConfig.master` {#RunConfig.master} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.num_ps_replicas` {#RunConfig.num_ps_replicas} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.task_id` {#RunConfig.task_id} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.task_type` {#RunConfig.task_type} diff --git a/tensorflow/g3doc/api_docs/python/control_flow_ops.md b/tensorflow/g3doc/api_docs/python/control_flow_ops.md index c22e5f1179..2f22520c8f 100644 --- a/tensorflow/g3doc/api_docs/python/control_flow_ops.md +++ b/tensorflow/g3doc/api_docs/python/control_flow_ops.md @@ -21,12 +21,12 @@ Return a tensor with the same shape and contents as the input tensor or value. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -121,7 +121,7 @@ Increments 'ref' until it reaches 'limit'. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `int32`, `int64`. Should be from a scalar `Variable` node. * <b>`limit`</b>: An `int`. If incrementing ref would bring it above limit, instead generates an @@ -130,7 +130,7 @@ Increments 'ref' until it reaches 'limit'. ##### Returns: - A `Tensor`. Has the same type as `ref`. + A `Output`. Has the same type as `ref`. A copy of the input before increment. If nothing else modifies the input, the values produced will all be distinct. @@ -408,13 +408,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -426,12 +426,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -446,13 +446,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -480,13 +480,13 @@ Returns the truth value of (x == y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -501,13 +501,13 @@ Returns the truth value of (x != y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -522,13 +522,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -543,13 +543,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -564,13 +564,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -585,13 +585,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -643,16 +643,16 @@ select(condition, t, e) ==> [[1, 2], ##### Args: -* <b>`condition`</b>: A `Tensor` of type `bool`. -* <b>`t`</b>: A `Tensor` which may have the same shape as `condition`. +* <b>`condition`</b>: An `Output` of type `bool`. +* <b>`t`</b>: An `Output` which may have the same shape as `condition`. If `condition` is rank 1, `t` may have higher rank, but its first dimension must match the size of `condition`. -* <b>`e`</b>: A `Tensor` with the same type and shape as `t`. +* <b>`e`</b>: An `Output` with the same type and shape as `t`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` with the same type and shape as `t` and `e`. + An `Output` with the same type and shape as `t` and `e`. - - - @@ -724,12 +724,12 @@ Equivalent to np.isfinite ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -745,12 +745,12 @@ Equivalent to np.isinf ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -766,12 +766,12 @@ Equivalent to np.isnan ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -804,13 +804,13 @@ that are not a number (NaN) or infinity (Inf). Otherwise, passes `tensor` as-is. ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`message`</b>: A `string`. Prefix of the error message. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. + A `Output`. Has the same type as `tensor`. - - - diff --git a/tensorflow/g3doc/api_docs/python/framework.md b/tensorflow/g3doc/api_docs/python/framework.md index cc1d27fead..8ad672deb1 100644 --- a/tensorflow/g3doc/api_docs/python/framework.md +++ b/tensorflow/g3doc/api_docs/python/framework.md @@ -1344,7 +1344,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -1372,13 +1373,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1393,13 +1394,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1440,13 +1441,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1503,13 +1504,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1584,13 +1585,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1629,12 +1630,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1666,13 +1667,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1687,13 +1688,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1708,13 +1709,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1735,12 +1736,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1768,13 +1769,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1818,13 +1819,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1839,13 +1840,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -1860,13 +1861,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1923,13 +1924,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1951,13 +1952,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2001,13 +2002,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2070,13 +2071,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2344,7 +2345,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -2372,13 +2374,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2393,13 +2395,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2440,13 +2442,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2503,13 +2505,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2584,13 +2586,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2629,12 +2631,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2666,13 +2668,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2687,13 +2689,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2708,13 +2710,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2735,12 +2737,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2768,13 +2770,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2818,13 +2820,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2839,13 +2841,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -2860,13 +2862,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2923,13 +2925,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -2951,13 +2953,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -3001,13 +3003,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -3070,13 +3072,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.cholesky.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.cholesky.md index 046c443925..2c5ae7f518 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.cholesky.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.cholesky.md @@ -10,11 +10,11 @@ containing the Cholesky decompositions for all input submatrices `[..., :, :]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[..., M, M]`. + A `Output`. Has the same type as `input`. Shape is `[..., M, M]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.depth_to_space.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.depth_to_space.md index 03dc6bb3b0..11c9f8d601 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.depth_to_space.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.depth_to_space.md @@ -84,12 +84,12 @@ x = [[ [1], [2], [5], [6]], ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`block_size`</b>: An `int` that is `>= 2`. The size of the spatial block, same as in Space2Depth. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_gradient.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_gradient.md index b363afe7ce..66228c261a 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_gradient.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_gradient.md @@ -5,23 +5,23 @@ Compute gradients for a FakeQuantWithMinMaxVars operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxVars operation. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxVars operation. min, max: Quantization interval, scalar floats. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). + A tuple of `Output` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). -* <b>`backprops_wrt_input`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. inputs: +* <b>`backprops_wrt_input`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. inputs: `gradients * (inputs >= min && inputs <= max)`. -* <b>`backprop_wrt_min`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. min parameter: +* <b>`backprop_wrt_min`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. min parameter: `sum(gradients * (inputs < min))`. -* <b>`backprop_wrt_max`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. max parameter: +* <b>`backprop_wrt_max`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. max parameter: `sum(gradients * (inputs > max))`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_per_channel_gradient.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_per_channel_gradient.md index a7a62e29b3..437f518819 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_per_channel_gradient.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fake_quant_with_min_max_vars_per_channel_gradient.md @@ -5,26 +5,26 @@ Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxVars operation, shape one of: `[d]`, `[b, d]`, `[b, h, w, d]`. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxVars operation, shape same as `gradients`. min, max: Quantization interval, floats of shape `[d]`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). + A tuple of `Output` objects (backprops_wrt_input, backprop_wrt_min, backprop_wrt_max). -* <b>`backprops_wrt_input`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. inputs, shape same as +* <b>`backprops_wrt_input`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. inputs, shape same as `inputs`: `gradients * (inputs >= min && inputs <= max)`. -* <b>`backprop_wrt_min`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. min parameter, shape `[d]`: +* <b>`backprop_wrt_min`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. min parameter, shape `[d]`: `sum_per_d(gradients * (inputs < min))`. -* <b>`backprop_wrt_max`</b>: A `Tensor` of type `float32`. Backpropagated gradients w.r.t. max parameter, shape `[d]`: +* <b>`backprop_wrt_max`</b>: An `Output` of type `float32`. Backpropagated gradients w.r.t. max parameter, shape `[d]`: `sum_per_d(gradients * (inputs > max))`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft.md index da37dd4933..fc8f898e50 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft.md @@ -7,12 +7,12 @@ dimension of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most dimension of `input` is replaced with its 1D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft2d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft2d.md index 81b83df8bb..b07c29d8e3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft2d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.fft2d.md @@ -7,12 +7,12 @@ Compute the 2-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 2 dimensions of `input` are replaced with their 2D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.floormod.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.floormod.md index 5ebd691835..75b72f4b45 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.floormod.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.floormod.md @@ -11,11 +11,11 @@ with a flooring divide. E.g. `floor(x / y) * y + mod(x, y) = x`. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.ifft3d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.ifft3d.md index 7d106f24a8..4e4c992db4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.ifft3d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.ifft3d.md @@ -7,12 +7,12 @@ Compute the inverse 3-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 3 dimensions of `input` are replaced with their inverse 3D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.image.decode_jpeg.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.image.decode_jpeg.md index f4c6f1340a..7fd0b4f9ed 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.image.decode_jpeg.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.image.decode_jpeg.md @@ -21,7 +21,7 @@ downscaling the image later. ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The JPEG-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The JPEG-encoded image. * <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for the decoded image. * <b>`ratio`</b>: An optional `int`. Defaults to `1`. Downscaling ratio. @@ -37,5 +37,5 @@ downscaling the image later. ##### Returns: - A `Tensor` of type `uint8`. 3-D with shape `[height, width, channels]`.. + An `Output` of type `uint8`. 3-D with shape `[height, width, channels]`.. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_finite.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_finite.md index 15d5a7df94..ecc26ddab8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_finite.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_finite.md @@ -9,10 +9,10 @@ Equivalent to np.isfinite ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_nan.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_nan.md index b1fd8de13c..74bafe9043 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_nan.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.is_nan.md @@ -9,10 +9,10 @@ Equivalent to np.isnan ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mod.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mod.md index 86978890b5..707cf96d99 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mod.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mod.md @@ -8,11 +8,11 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mul.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mul.md index 1ac6abe73a..f7fcbaa598 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mul.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.mul.md @@ -8,11 +8,11 @@ Returns x * y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.nn.avg_pool3d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.nn.avg_pool3d.md index 5bb4dcf68f..c5895eb0c7 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.nn.avg_pool3d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.nn.avg_pool3d.md @@ -5,7 +5,7 @@ Performs 3D average pooling on the input. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, depth, rows, cols, channels]` tensor to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 5`. 1-D tensor of length 5. The size of the window for each dimension of @@ -19,6 +19,6 @@ Performs 3D average pooling on the input. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The average pooled output tensor. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.not_equal.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.not_equal.md index 5ed8df49d5..9879c14648 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.not_equal.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.not_equal.md @@ -8,11 +8,11 @@ Returns the truth value of (x != y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.quantize_v2.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.quantize_v2.md index a02df53efe..4b3595b6f6 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.quantize_v2.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.quantize_v2.md @@ -55,10 +55,10 @@ operations that have to perform further calculations on them. ##### Args: -* <b>`input`</b>: A `Tensor` of type `float32`. -* <b>`min_range`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: An `Output` of type `float32`. +* <b>`min_range`</b>: An `Output` of type `float32`. The minimum scalar value possibly produced for the input. -* <b>`max_range`</b>: A `Tensor` of type `float32`. +* <b>`max_range`</b>: An `Output` of type `float32`. The maximum scalar value possibly produced for the input. * <b>`T`</b>: A `tf.DType` from: `tf.qint8, tf.quint8, tf.qint16, tf.quint16, tf.qint32`. * <b>`mode`</b>: An optional `string` from: `"MIN_COMBINED", "MIN_FIRST"`. Defaults to `"MIN_COMBINED"`. @@ -66,9 +66,9 @@ operations that have to perform further calculations on them. ##### Returns: - A tuple of `Tensor` objects (output, output_min, output_max). + A tuple of `Output` objects (output, output_min, output_max). -* <b>`output`</b>: A `Tensor` of type `T`. The quantized data produced from the float input. -* <b>`output_min`</b>: A `Tensor` of type `float32`. The actual minimum scalar value used for the output. -* <b>`output_max`</b>: A `Tensor` of type `float32`. The actual maximum scalar value used for the output. +* <b>`output`</b>: A `Output` of type `T`. The quantized data produced from the float input. +* <b>`output_min`</b>: An `Output` of type `float32`. The actual minimum scalar value used for the output. +* <b>`output_max`</b>: An `Output` of type `float32`. The actual maximum scalar value used for the output. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reshape.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reshape.md index 05de3a2779..cdf46c3e67 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reshape.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reshape.md @@ -62,12 +62,12 @@ reshape(t, []) ==> 7 ##### Args: -* <b>`tensor`</b>: A `Tensor`. -* <b>`shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`tensor`</b>: A `Output`. +* <b>`shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Defines the shape of the output tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. + A `Output`. Has the same type as `tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reverse_sequence.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reverse_sequence.md index 03dd068320..e5fc6dc419 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reverse_sequence.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.reverse_sequence.md @@ -60,8 +60,8 @@ output[2:, :, 3, :, ...] = input[2:, :, 3, :, ...] ##### Args: -* <b>`input`</b>: A `Tensor`. The input to reverse. -* <b>`seq_lengths`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. The input to reverse. +* <b>`seq_lengths`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with length `input.dims(batch_dim)` and `max(seq_lengths) < input.dims(seq_dim)` * <b>`seq_dim`</b>: An `int`. The dimension which is partially reversed. @@ -71,6 +71,6 @@ output[2:, :, 3, :, ...] = input[2:, :, 3, :, ...] ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The partially reversed input. It has the same shape as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.segment_min.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.segment_min.md index 5cacf2cf72..28cc205863 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.segment_min.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.segment_min.md @@ -17,15 +17,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.Tensor.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.Tensor.md index 910a5c6784..2f1de8df15 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.Tensor.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.Tensor.md @@ -210,7 +210,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -238,13 +239,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -259,13 +260,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -306,13 +307,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -369,13 +370,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -450,13 +451,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -495,12 +496,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -532,13 +533,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -553,13 +554,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -574,13 +575,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -601,12 +602,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -634,13 +635,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -684,13 +685,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -705,13 +706,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -726,13 +727,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -789,13 +790,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -817,13 +818,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -867,13 +868,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -936,13 +937,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.assign.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.assign.md index f72385be60..804fd300e8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.assign.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.assign.md @@ -8,9 +8,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. +* <b>`ref`</b>: A mutable `Output`. Should be from a `Variable` node. May be uninitialized. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be assigned to the variable. * <b>`validate_shape`</b>: An optional `bool`. Defaults to `True`. If true, the operation will validate that the shape diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.batch_to_space.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.batch_to_space.md index a5ecd15aec..42b0b92952 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.batch_to_space.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.batch_to_space.md @@ -13,11 +13,11 @@ followed by cropping along the `height` and `width` dimensions. ##### Args: -* <b>`input`</b>: A `Tensor`. 4-D tensor with shape +* <b>`input`</b>: A `Output`. 4-D tensor with shape `[batch*block_size*block_size, height_pad/block_size, width_pad/block_size, depth]`. Note that the batch size of the input tensor must be divisible by `block_size * block_size`. -* <b>`crops`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`crops`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies how many elements to crop from the intermediate result across the spatial dimensions as follows: @@ -29,7 +29,7 @@ followed by cropping along the `height` and `width` dimensions. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. 4-D with shape `[batch, height, width, depth]`, where: height = height_pad - crop_top - crop_bottom diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.diag.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.diag.md index 3279122875..33f62c3ee7 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.diag.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.diag.md @@ -23,11 +23,11 @@ tf.diag(diagonal) ==> [[1, 0, 0, 0] ##### Args: -* <b>`diagonal`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`diagonal`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. Rank k tensor where k is at most 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `diagonal`. + A `Output`. Has the same type as `diagonal`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.fake_quant_with_min_max_args_gradient.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.fake_quant_with_min_max_args_gradient.md index 5c93c3e046..31f87cd0e5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.fake_quant_with_min_max_args_gradient.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.fake_quant_with_min_max_args_gradient.md @@ -5,9 +5,9 @@ Compute gradients for a FakeQuantWithMinMaxArgs operation. ##### Args: -* <b>`gradients`</b>: A `Tensor` of type `float32`. +* <b>`gradients`</b>: An `Output` of type `float32`. Backpropagated gradients above the FakeQuantWithMinMaxArgs operation. -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. Values passed as inputs to the FakeQuantWithMinMaxArgs operation. * <b>`min`</b>: An optional `float`. Defaults to `-6`. * <b>`max`</b>: An optional `float`. Defaults to `6`. @@ -15,7 +15,7 @@ Compute gradients for a FakeQuantWithMinMaxArgs operation. ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. Backpropagated gradients below the FakeQuantWithMinMaxArgs operation: `gradients * (inputs >= min && inputs <= max)`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.greater_equal.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.greater_equal.md index d6ce057c13..11437eea06 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.greater_equal.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.greater_equal.md @@ -8,11 +8,11 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.igammac.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.igammac.md index 2d935bb6e3..008dd1690d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.igammac.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.igammac.md @@ -19,11 +19,11 @@ Gamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.image.resize_bicubic.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.image.resize_bicubic.md index 1805c7423d..8ab375d91f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.image.resize_bicubic.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.image.resize_bicubic.md @@ -7,7 +7,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -19,6 +19,6 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.invert_permutation.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.invert_permutation.md index 20cab18208..a2fbc832b8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.invert_permutation.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.invert_permutation.md @@ -21,10 +21,10 @@ invert_permutation(x) ==> [2, 4, 3, 0, 1] ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. 1-D. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. 1-D. + A `Output`. Has the same type as `x`. 1-D. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.linspace.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.linspace.md index 29b8993fe6..e79dfd2f4d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.linspace.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.linspace.md @@ -15,15 +15,15 @@ tf.linspace(10.0, 12.0, 3, name="linspace") => [ 10.0 11.0 12.0] ##### Args: -* <b>`start`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`start`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. First entry in the range. -* <b>`stop`</b>: A `Tensor`. Must have the same type as `start`. +* <b>`stop`</b>: A `Output`. Must have the same type as `start`. Last entry in the range. -* <b>`num`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`num`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Number of values to generate. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `start`. 1-D. The generated values. + A `Output`. Has the same type as `start`. 1-D. The generated values. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.log1p.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.log1p.md index e861034528..a6609c73e2 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.log1p.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.log1p.md @@ -7,10 +7,10 @@ I.e., \\(y = \log_e (1 + x)\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.matching_files.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.matching_files.md index 297462d580..16f2f21f25 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.matching_files.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.matching_files.md @@ -8,10 +8,10 @@ basename portion of the pattern, not in the directory portion. ##### Args: -* <b>`pattern`</b>: A `Tensor` of type `string`. A (scalar) shell wildcard pattern. +* <b>`pattern`</b>: An `Output` of type `string`. A (scalar) shell wildcard pattern. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. A vector of matching filenames. + An `Output` of type `string`. A vector of matching filenames. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.negative.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.negative.md index 2ebfb528b0..c8ee5a2455 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.negative.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.negative.md @@ -7,10 +7,10 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.read_file.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.read_file.md index 3c0ad3652a..d15c39f539 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.read_file.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.read_file.md @@ -5,10 +5,10 @@ Reads and outputs the entire contents of the input filename. ##### Args: -* <b>`filename`</b>: A `Tensor` of type `string`. +* <b>`filename`</b>: An `Output` of type `string`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.reduce_join.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.reduce_join.md index ad49e98274..6498703421 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.reduce_join.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.reduce_join.md @@ -29,9 +29,9 @@ tf.reduce_join(a, []) ==> ["abcd"] ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `string`. +* <b>`inputs`</b>: An `Output` of type `string`. The input to be joined. All reduced indices must have non-zero size. -* <b>`reduction_indices`</b>: A `Tensor` of type `int32`. +* <b>`reduction_indices`</b>: An `Output` of type `int32`. The dimensions to reduce over. Dimensions are reduced in the order specified. Omitting `reduction_indices` is equivalent to passing `[n-1, n-2, ..., 0]`. Negative indices from `-n` to `-1` are supported. @@ -43,7 +43,7 @@ tf.reduce_join(a, []) ==> ["abcd"] ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. Has shape equal to that of the input with reduced dimensions removed or set to `1` depending on `keep_dims`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.string_to_hash_bucket_strong.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.string_to_hash_bucket_strong.md index 764dfe8431..6db43b6c55 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.string_to_hash_bucket_strong.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.string_to_hash_bucket_strong.md @@ -16,7 +16,7 @@ time than `tf.string_to_hash_bucket_fast`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. The strings to assign a hash bucket. +* <b>`input`</b>: An `Output` of type `string`. The strings to assign a hash bucket. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`key`</b>: A list of `ints`. The key for the keyed hash function passed as a list of two uint64 @@ -25,6 +25,6 @@ time than `tf.string_to_hash_bucket_fast`. ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.betainc.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.betainc.md index 9da04a3642..18ea482983 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.betainc.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.betainc.md @@ -19,12 +19,12 @@ beta function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`b`</b>: A `Tensor`. Must have the same type as `a`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`b`</b>: A `Output`. Must have the same type as `a`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.digamma.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.digamma.md index 8729e7ecfe..d415f0828c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.digamma.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.digamma.md @@ -7,10 +7,10 @@ Computes Psi, the derivative of Lgamma (the log of the absolute value of ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.encode_base64.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.encode_base64.md index 20fef36bcb..33798774f5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.encode_base64.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.encode_base64.md @@ -12,12 +12,12 @@ Web-safe means that the encoder uses - and _ instead of + and /. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Strings to be encoded. +* <b>`input`</b>: An `Output` of type `string`. Strings to be encoded. * <b>`pad`</b>: An optional `bool`. Defaults to `False`. Bool whether padding is applied at the ends. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Input strings encoded in base64. + An `Output` of type `string`. Input strings encoded in base64. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.expand_dims.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.expand_dims.md index 5e8c0c8f7a..483e7bf7dd 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.expand_dims.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.expand_dims.md @@ -36,15 +36,15 @@ size 1. ##### Args: -* <b>`input`</b>: A `Tensor`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 0-D (scalar). Specifies the axisension index at which to expand the shape of `input`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Contains the same data as `input`, but its shape has an additional axisension of size 1 added. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.floor_div.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.floor_div.md index da18338be6..992311c3a3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.floor_div.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.floor_div.md @@ -8,11 +8,11 @@ Returns x // y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.gather_nd.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.gather_nd.md index 22cccc9d9a..a008eaae94 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.gather_nd.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.gather_nd.md @@ -97,14 +97,14 @@ Batched indexing into a 3-tensor: ##### Args: -* <b>`params`</b>: A `Tensor`. `P-D`. The tensor from which to gather values. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`params`</b>: A `Output`. `P-D`. The tensor from which to gather values. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. `Q-D`. Index tensor having shape `[d_0, ..., d_{Q-2}, K]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `params`. + A `Output`. Has the same type as `params`. `(P+Q-K-1)-D`. Values from `params` gathered from indices given by `indices`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.identity.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.identity.md index 13f1318601..4856fbfd34 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.identity.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.identity.md @@ -5,10 +5,10 @@ Return a tensor with the same shape and contents as the input tensor or value. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.imag.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.imag.md index 3df72723cf..e6a0ed1a39 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.imag.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.imag.md @@ -18,7 +18,8 @@ tf.imag(input) ==> [4.75, 5.75] ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `complex64`, `complex128`. +* <b>`input`</b>: A `Tensor`. Must be one of the following types: `complex64`, + `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.crop_and_resize.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.crop_and_resize.md index aace65153a..492a52de15 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.crop_and_resize.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.crop_and_resize.md @@ -14,10 +14,10 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. ##### Args: -* <b>`image`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`image`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. A 4-D tensor of shape `[batch, image_height, image_width, depth]`. Both `image_height` and `image_width` need to be positive. -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor specifies the coordinates of a box in the `box_ind[i]` image and is specified in normalized coordinates `[y1, x1, y2, x2]`. A normalized coordinate value of @@ -28,10 +28,10 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. image. The width dimension is treated similarly. Normalized coordinates outside the `[0, 1]` range are allowed, in which case we use `extrapolation_value` to extrapolate the input image values. -* <b>`box_ind`</b>: A `Tensor` of type `int32`. +* <b>`box_ind`</b>: An `Output` of type `int32`. A 1-D tensor of shape `[num_boxes]` with int32 values in `[0, batch)`. The value of `box_ind[i]` specifies the image that the `i`-th box refers to. -* <b>`crop_size`</b>: A `Tensor` of type `int32`. +* <b>`crop_size`</b>: An `Output` of type `int32`. A 1-D tensor of 2 elements, `size = [crop_height, crop_width]`. All cropped image patches are resized to this size. The aspect ratio of the image content is not preserved. Both `crop_height` and `crop_width` need to be @@ -45,6 +45,6 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. A 4-D tensor of shape `[num_boxes, crop_height, crop_width, depth]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.encode_jpeg.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.encode_jpeg.md index 24b1886c10..d053fe2c30 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.encode_jpeg.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.image.encode_jpeg.md @@ -22,7 +22,7 @@ in function of the number of channels in `image`: ##### Args: -* <b>`image`</b>: A `Tensor` of type `uint8`. +* <b>`image`</b>: An `Output` of type `uint8`. 3-D with shape `[height, width, channels]`. * <b>`format`</b>: An optional `string` from: `"", "grayscale", "rgb"`. Defaults to `""`. Per pixel image format. @@ -47,5 +47,5 @@ in function of the number of channels in `image`: ##### Returns: - A `Tensor` of type `string`. 0-D. JPEG-encoded image. + An `Output` of type `string`. 0-D. JPEG-encoded image. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.matmul.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.matmul.md index ec289a77d4..d20ff3a5ba 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.matmul.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.matmul.md @@ -1,19 +1,22 @@ -### `tf.matmul(a, b, transpose_a=False, transpose_b=False, a_is_sparse=False, b_is_sparse=False, name=None)` {#matmul} +### `tf.matmul(a, b, transpose_a=False, transpose_b=False, adjoint_a=False, adjoint_b=False, a_is_sparse=False, b_is_sparse=False, name=None)` {#matmul} Multiplies matrix `a` by matrix `b`, producing `a` * `b`. -The inputs must be two-dimensional matrices, with matching inner dimensions, -possibly after transposition. +The inputs must be matrices (or tensors of rank > 2, representing batches of +matrices), with matching inner dimensions, possibly after transposition. Both matrices must be of the same type. The supported types are: -`float32`, `float64`, `int32`, `complex64`. +`float16`, `float32`, `float64`, `int32`, `complex64`, `complex128`. -Either matrix can be transposed on the fly by setting the corresponding flag -to `True`. This is `False` by default. +Either matrix can be transposed or adjointed (conjugated and transposed) on +the fly by setting one of the corresponding flag to `True`. These are `False` +by default. If one or both of the matrices contain a lot of zeros, a more efficient multiplication algorithm can be used by setting the corresponding `a_is_sparse` or `b_is_sparse` flag to `True`. These are `False` by default. +This optimization is only available for plain matrices (rank-2 tensors) with +datatypes `bfloat16` or `float32`. For example: @@ -27,20 +30,55 @@ b = tf.constant([7, 8, 9, 10, 11, 12], shape=[3, 2]) => [[7. 8.] [11. 12.]] c = tf.matmul(a, b) => [[58 64] [139 154]] + + +# 3-D tensor `a` +a = tf.constant(np.arange(1,13), shape=[2, 2, 3]) => [[[ 1. 2. 3.] + [ 4. 5. 6.]], + [[ 7. 8. 9.] + [10. 11. 12.]]] + +# 3-D tensor `b` +b = tf.constant(np.arange(13,25), shape=[2, 3, 2]) => [[[13. 14.] + [15. 16.] + [17. 18.]], + [[19. 20.] + [21. 22.] + [23. 24.]]] +c = tf.matmul(a, b) => [[[ 94 100] + [229 244]], + [[508 532] + [697 730]]] ``` ##### Args: -* <b>`a`</b>: `Tensor` of type `float32`, `float64`, `int32` or `complex64`. -* <b>`b`</b>: `Tensor` with same type as `a`. +* <b>`a`</b>: `Tensor` of type `float16`, `float32`, `float64`, `int32`, `complex64`, + `complex128` and rank > 1. +* <b>`b`</b>: `Tensor` with same type and rank as `a`. * <b>`transpose_a`</b>: If `True`, `a` is transposed before multiplication. * <b>`transpose_b`</b>: If `True`, `b` is transposed before multiplication. +* <b>`adjoint_a`</b>: If `True`, `a` is conjugated and transposed before + multiplication. +* <b>`adjoint_b`</b>: If `True`, `b` is conjugated and transposed before + multiplication. * <b>`a_is_sparse`</b>: If `True`, `a` is treated as a sparse matrix. * <b>`b_is_sparse`</b>: If `True`, `b` is treated as a sparse matrix. * <b>`name`</b>: Name for the operation (optional). ##### Returns: - A `Tensor` of the same type as `a`. + A `Tensor` of the same type as `a` and `b` where each inner-most matrix is + the product of the corresponding matrices in `a` and `b, e.g. if all + transpose or adjoint attributes are `False`: + + output[..., :, :] = a[..., :, :] * b[..., :, :] , + + +##### Raises: + + +* <b>`ValueError`</b>: If transpose_a and adjoint_a, or transpose_b and adjoint_b + are both set to True. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.minimum.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.minimum.md index 9bcd03f6e7..2c1dd7e9a5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.minimum.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.minimum.md @@ -8,11 +8,11 @@ Returns the min of x and y (i.e. x < y ? x : y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.relu.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.relu.md index 5811a1da96..55c8ae1730 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.relu.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.relu.md @@ -5,10 +5,10 @@ Computes rectified linear: `max(features, 0)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_nd_sub.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_nd_sub.md index 1d16c8e06c..c9c186784f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_nd_sub.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_nd_sub.md @@ -4,7 +4,7 @@ Applies sparse subtraction between `updates` and individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -13,7 +13,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -39,12 +39,12 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to subtract from ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. @@ -55,7 +55,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_update.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_update.md index f865b8e9e8..d51e9979ed 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_update.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_update.md @@ -29,10 +29,10 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. Should be from a `Variable` node. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to store in `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `True`. If True, the assignment will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.setdiff1d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.setdiff1d.md index 3bd95f13c5..2de3197eae 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.setdiff1d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.setdiff1d.md @@ -27,15 +27,15 @@ idx ==> [1, 3, 5] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. Values to keep. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. 1-D. Values to remove. +* <b>`x`</b>: A `Output`. 1-D. Values to keep. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. 1-D. Values to remove. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (out, idx). + A tuple of `Output` objects (out, idx). -* <b>`out`</b>: A `Tensor`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. +* <b>`out`</b>: A `Output`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.shape_n.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.shape_n.md index 5a5eca2762..0d13cf4f72 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.shape_n.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.shape_n.md @@ -7,11 +7,11 @@ This operation returns N 1-D integer tensors representing shape of `input[i]s`. ##### Args: -* <b>`input`</b>: A list of at least 1 `Tensor` objects of the same type. +* <b>`input`</b>: A list of at least 1 `Output` objects of the same type. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A list with the same number of `Tensor` objects as `input` of `Tensor` objects of type out_type. + A list with the same number of `Output` objects as `input` of `Output` objects of type out_type. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.sin.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.sin.md index f69c58bee0..44724d2ef4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.sin.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.sin.md @@ -5,10 +5,10 @@ Computes sin of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.space_to_batch_nd.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.space_to_batch_nd.md index 86a441528a..d3a79e1fae 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.space_to_batch_nd.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.space_to_batch_nd.md @@ -14,12 +14,12 @@ precise description. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`, where spatial_shape has `M` dimensions. -* <b>`block_shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`block_shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with shape `[M]`, all values must be >= 1. -* <b>`paddings`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`paddings`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D with shape `[M, 2]`, all values must be >= 0. `paddings[i] = [pad_start, pad_end]` specifies the padding for input dimension `i + 1`, which corresponds to spatial dimension `i`. It is required that @@ -133,5 +133,5 @@ precise description. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.write_file.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.write_file.md index ccccf9b43b..99a2f250fd 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.write_file.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.write_file.md @@ -5,9 +5,9 @@ Writes contents to the file at input filename. Creates file if not existing. ##### Args: -* <b>`filename`</b>: A `Tensor` of type `string`. +* <b>`filename`</b>: An `Output` of type `string`. scalar. The name of the file to which we write the contents. -* <b>`contents`</b>: A `Tensor` of type `string`. +* <b>`contents`</b>: An `Output` of type `string`. scalar. The content to be written to the output file. * <b>`name`</b>: A name for the operation (optional). diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.SparseTensor.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.SparseTensor.md index d89b4e70c4..d2126122c6 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.SparseTensor.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.SparseTensor.md @@ -149,20 +149,20 @@ the other direction. ##### Args: -* <b>`sp_indices`</b>: A `Tensor` of type `int64`. +* <b>`sp_indices`</b>: An `Output` of type `int64`. 2-D. `N x R` matrix with the indices of non-empty values in a SparseTensor, possibly not in canonical ordering. -* <b>`sp_values`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`sp_values`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 1-D. `N` non-empty values corresponding to `sp_indices`. -* <b>`sp_shape`</b>: A `Tensor` of type `int64`. +* <b>`sp_shape`</b>: An `Output` of type `int64`. 1-D. Shape of the input SparseTensor. -* <b>`dense`</b>: A `Tensor`. Must have the same type as `sp_values`. +* <b>`dense`</b>: A `Output`. Must have the same type as `sp_values`. `R`-D. The dense Tensor operand. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `sp_values`. + A `Output`. Has the same type as `sp_values`. 1-D. The `N` values that are operated on. @@ -182,20 +182,20 @@ the other direction. ##### Args: -* <b>`sp_indices`</b>: A `Tensor` of type `int64`. +* <b>`sp_indices`</b>: An `Output` of type `int64`. 2-D. `N x R` matrix with the indices of non-empty values in a SparseTensor, possibly not in canonical ordering. -* <b>`sp_values`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`sp_values`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 1-D. `N` non-empty values corresponding to `sp_indices`. -* <b>`sp_shape`</b>: A `Tensor` of type `int64`. +* <b>`sp_shape`</b>: An `Output` of type `int64`. 1-D. Shape of the input SparseTensor. -* <b>`dense`</b>: A `Tensor`. Must have the same type as `sp_values`. +* <b>`dense`</b>: A `Output`. Must have the same type as `sp_values`. `R`-D. The dense Tensor operand. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `sp_values`. + A `Output`. Has the same type as `sp_values`. 1-D. The `N` values that are operated on. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ceil.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ceil.md index 34e4a7feed..ec398efff8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ceil.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ceil.md @@ -5,10 +5,10 @@ Returns element-wise smallest integer in not less than x. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.check_numerics.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.check_numerics.md index 46a8f6f7db..98384be7df 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.check_numerics.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.check_numerics.md @@ -8,11 +8,11 @@ that are not a number (NaN) or infinity (Inf). Otherwise, passes `tensor` as-is. ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`message`</b>: A `string`. Prefix of the error message. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. + A `Output`. Has the same type as `tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_csv.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_csv.md index f2ebf6945b..b536c0d7ee 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_csv.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_csv.md @@ -9,10 +9,10 @@ Note that we allow leading and trailing spaces with int or float field. ##### Args: -* <b>`records`</b>: A `Tensor` of type `string`. +* <b>`records`</b>: An `Output` of type `string`. Each string is a record/row in the csv and all records should have the same format. -* <b>`record_defaults`</b>: A list of `Tensor` objects with types from: `float32`, `int32`, `int64`, `string`. +* <b>`record_defaults`</b>: A list of `Output` objects with types from: `float32`, `int32`, `int64`, `string`. One tensor per column of the input record, with either a scalar default value for that column or empty if the column is required. * <b>`field_delim`</b>: An optional `string`. Defaults to `","`. @@ -21,6 +21,6 @@ Note that we allow leading and trailing spaces with int or float field. ##### Returns: - A list of `Tensor` objects. Has the same type as `record_defaults`. + A list of `Output` objects. Has the same type as `record_defaults`. Each tensor will have the same shape as records. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_raw.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_raw.md index 8beeae4c00..8d6231d7d4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_raw.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.decode_raw.md @@ -5,7 +5,7 @@ Reinterpret the bytes of a string as a vector of numbers. ##### Args: -* <b>`bytes`</b>: A `Tensor` of type `string`. +* <b>`bytes`</b>: An `Output` of type `string`. All the elements must have the same length. * <b>`out_type`</b>: A `tf.DType` from: `tf.half, tf.float32, tf.float64, tf.int32, tf.uint8, tf.int16, tf.int8, tf.int64`. * <b>`little_endian`</b>: An optional `bool`. Defaults to `True`. @@ -16,7 +16,7 @@ Reinterpret the bytes of a string as a vector of numbers. ##### Returns: - A `Tensor` of type `out_type`. + A `Output` of type `out_type`. A Tensor with one more dimension than the input `bytes`. The added dimension will have size equal to the length of the elements of `bytes` divided by the number of bytes to represent `out_type`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.exp.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.exp.md index f31531e762..540f5260d0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.exp.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.exp.md @@ -5,10 +5,10 @@ Computes exponential of x element-wise. \\(y = e^x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars.md index d7815f0414..6150ed8a8b 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars.md @@ -14,12 +14,12 @@ This operation has a gradient and thus allows for training `min` and `max` value ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars_per_channel.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars_per_channel.md index bc39cf9570..05bfdd1d34 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars_per_channel.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.fake_quant_with_min_max_vars_per_channel.md @@ -14,12 +14,12 @@ This operation has a gradient and thus allows for training `min` and `max` value ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. -* <b>`min`</b>: A `Tensor` of type `float32`. -* <b>`max`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. +* <b>`min`</b>: An `Output` of type `float32`. +* <b>`max`</b>: An `Output` of type `float32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ifft2d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ifft2d.md index d19b164d8c..73889f4b0e 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ifft2d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.ifft2d.md @@ -7,12 +7,12 @@ Compute the inverse 2-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 2 dimensions of `input` are replaced with their inverse 2D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.image.decode_png.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.image.decode_png.md index 4332af7704..2de8847af7 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.image.decode_png.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.image.decode_png.md @@ -18,7 +18,7 @@ of color channels. ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The PNG-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The PNG-encoded image. * <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for the decoded image. * <b>`dtype`</b>: An optional `tf.DType` from: `tf.uint8, tf.uint16`. Defaults to `tf.uint8`. @@ -26,5 +26,5 @@ of color channels. ##### Returns: - A `Tensor` of type `dtype`. 3-D with shape `[height, width, channels]`. + A `Output` of type `dtype`. 3-D with shape `[height, width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.fractional_max_pool.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.fractional_max_pool.md index 8f8fb0237c..286d1a0c64 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.fractional_max_pool.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.fractional_max_pool.md @@ -34,7 +34,7 @@ For more details on fractional max pooling, see this paper: ##### Args: -* <b>`value`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. +* <b>`value`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. 4-D with shape `[batch, height, width, channels]`. * <b>`pooling_ratio`</b>: A list of `floats` that has length `>= 4`. Pooling ratio for each dimension of `value`, currently only @@ -73,9 +73,9 @@ For more details on fractional max pooling, see this paper: ##### Returns: - A tuple of `Tensor` objects (output, row_pooling_sequence, col_pooling_sequence). + A tuple of `Output` objects (output, row_pooling_sequence, col_pooling_sequence). -* <b>`output`</b>: A `Tensor`. Has the same type as `value`. output tensor after fractional max pooling. -* <b>`row_pooling_sequence`</b>: A `Tensor` of type `int64`. row pooling sequence, needed to calculate gradient. -* <b>`col_pooling_sequence`</b>: A `Tensor` of type `int64`. column pooling sequence, needed to calculate gradient. +* <b>`output`</b>: A `Output`. Has the same type as `value`. output tensor after fractional max pooling. +* <b>`row_pooling_sequence`</b>: An `Output` of type `int64`. row pooling sequence, needed to calculate gradient. +* <b>`col_pooling_sequence`</b>: An `Output` of type `int64`. column pooling sequence, needed to calculate gradient. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_max_pool.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_max_pool.md index 3ddffd5b83..314c1d4dca 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_max_pool.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_max_pool.md @@ -5,11 +5,11 @@ Produces the max pool of the input tensor for quantized types. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. The 4D (batch x rows x cols x depth) Tensor to MaxReduce over. -* <b>`min_input`</b>: A `Tensor` of type `float32`. +* <b>`min_input`</b>: An `Output` of type `float32`. The float value that the lowest quantized input value represents. -* <b>`max_input`</b>: A `Tensor` of type `float32`. +* <b>`max_input`</b>: An `Output` of type `float32`. The float value that the highest quantized input value represents. * <b>`ksize`</b>: A list of `ints`. The size of the window for each dimension of the input tensor. @@ -23,9 +23,9 @@ Produces the max pool of the input tensor for quantized types. ##### Returns: - A tuple of `Tensor` objects (output, min_output, max_output). + A tuple of `Output` objects (output, min_output, max_output). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. -* <b>`min_output`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized output value represents. -* <b>`max_output`</b>: A `Tensor` of type `float32`. The float value that the highest quantized output value represents. +* <b>`output`</b>: A `Output`. Has the same type as `input`. +* <b>`min_output`</b>: An `Output` of type `float32`. The float value that the lowest quantized output value represents. +* <b>`max_output`</b>: An `Output` of type `float32`. The float value that the highest quantized output value represents. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_relu_x.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_relu_x.md index 2738a4bdab..05588c2f32 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_relu_x.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.quantized_relu_x.md @@ -5,20 +5,20 @@ Computes Quantized Rectified Linear X: `min(max(features, 0), max_value)` ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. -* <b>`max_value`</b>: A `Tensor` of type `float32`. -* <b>`min_features`</b>: A `Tensor` of type `float32`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`max_value`</b>: An `Output` of type `float32`. +* <b>`min_features`</b>: An `Output` of type `float32`. The float value that the lowest quantized value represents. -* <b>`max_features`</b>: A `Tensor` of type `float32`. +* <b>`max_features`</b>: An `Output` of type `float32`. The float value that the highest quantized value represents. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.qint8, tf.quint8, tf.qint16, tf.quint16, tf.qint32`. Defaults to `tf.quint8`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (activations, min_activations, max_activations). + A tuple of `Output` objects (activations, min_activations, max_activations). -* <b>`activations`</b>: A `Tensor` of type `out_type`. Has the same output shape as "features". -* <b>`min_activations`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized value represents. -* <b>`max_activations`</b>: A `Tensor` of type `float32`. The float value that the highest quantized value represents. +* <b>`activations`</b>: A `Output` of type `out_type`. Has the same output shape as "features". +* <b>`min_activations`</b>: An `Output` of type `float32`. The float value that the lowest quantized value represents. +* <b>`max_activations`</b>: An `Output` of type `float32`. The float value that the highest quantized value represents. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.softsign.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.softsign.md index 971b2a8134..563a0424b1 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.softsign.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.softsign.md @@ -5,10 +5,10 @@ Computes softsign: `features / (abs(features) + 1)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.parse_tensor.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.parse_tensor.md index 796eb39598..2b2a7bfa1d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.parse_tensor.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.parse_tensor.md @@ -5,7 +5,7 @@ Transforms a serialized tensorflow.TensorProto proto into a Tensor. ##### Args: -* <b>`serialized`</b>: A `Tensor` of type `string`. +* <b>`serialized`</b>: An `Output` of type `string`. A scalar string containing a serialized TensorProto proto. * <b>`out_type`</b>: A `tf.DType`. The type of the serialized tensor. The provided type must match the @@ -14,5 +14,5 @@ Transforms a serialized tensorflow.TensorProto proto into a Tensor. ##### Returns: - A `Tensor` of type `out_type`. A Tensor of type `out_type`. + A `Output` of type `out_type`. A Tensor of type `out_type`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_mul.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_mul.md index 94da4712d4..cbccbafba7 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_mul.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_mul.md @@ -24,11 +24,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to multiply to `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the operation will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_nd_add.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_nd_add.md index 4d1472205d..b67a5890b1 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_nd_add.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.scatter_nd_add.md @@ -4,7 +4,7 @@ Applies sparse addition between `updates` and individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -13,7 +13,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -39,12 +39,12 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to add to ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. @@ -55,7 +55,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.segment_mean.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.segment_mean.md index 5d901859a9..948a8864a2 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.segment_mean.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.segment_mean.md @@ -18,15 +18,15 @@ values summed. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.stop_gradient.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.stop_gradient.md index 53759f49ff..3afaaee416 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.stop_gradient.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.stop_gradient.md @@ -25,10 +25,10 @@ to pretend that the value was a constant. Some examples include: ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.substr.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.substr.md index 0f5a21cc14..862ae23bef 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.substr.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.substr.md @@ -1,8 +1,8 @@ ### `tf.substr(input, pos, len, name=None)` {#substr} -Return substrings from `Tensor` of strings. +Return substrings from `Output` of strings. -For each string in the input `Tensor`, creates a substring starting at index +For each string in the input `Output`, creates a substring starting at index `pos` with a total length of `len`. If `len` defines a substring that would extend beyond the length of the input @@ -79,14 +79,14 @@ output = [b'hir', b'ee', b'n"] ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Tensor of strings -* <b>`pos`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: An `Output` of type `string`. Tensor of strings +* <b>`pos`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Scalar defining the position of first character in each substring -* <b>`len`</b>: A `Tensor`. Must have the same type as `pos`. +* <b>`len`</b>: A `Output`. Must have the same type as `pos`. Scalar defining the number of characters to include in each substring * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Tensor of substrings + An `Output` of type `string`. Tensor of substrings diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.truncatediv.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.truncatediv.md index 99c9d55cea..dc59594c1c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.truncatediv.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.truncatediv.md @@ -13,11 +13,11 @@ Python Semantics. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.unique_with_counts.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.unique_with_counts.md index 0228699c63..75778e01bf 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.unique_with_counts.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.unique_with_counts.md @@ -23,15 +23,15 @@ count ==> [2, 1, 3, 1, 2] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. +* <b>`x`</b>: A `Output`. 1-D. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (y, idx, count). + A tuple of `Output` objects (y, idx, count). -* <b>`y`</b>: A `Tensor`. Has the same type as `x`. 1-D. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. -* <b>`count`</b>: A `Tensor` of type `out_idx`. 1-D. +* <b>`y`</b>: A `Output`. Has the same type as `x`. 1-D. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. +* <b>`count`</b>: A `Output` of type `out_idx`. 1-D. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.argmin.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.argmin.md index ae686a24cb..b9609883a0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.argmin.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.argmin.md @@ -5,13 +5,13 @@ Returns the index with the smallest value across axiss of a tensor. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. int32, 0 <= axis < rank(input). Describes which axis of the input Tensor to reduce across. For vectors, use axis = 0. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.learn.RunConfig.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.learn.RunConfig.md index 1da1181250..4e8ea0dd85 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.learn.RunConfig.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.learn.RunConfig.md @@ -1,65 +1,29 @@ -This class specifies the specific configurations for the run. +This class specifies the configurations for an `Estimator` run. -If you're a Google-internal user using command line flags with learn_runner.py -(for instance, to do distributed training or to use parameter servers), you -probably want to use learn_runner.EstimatorConfig instead. +If you're a Google-internal user using command line flags with +`learn_runner.py` (for instance, to do distributed training or to use +parameter servers), you probably want to use `learn_runner.EstimatorConfig` +instead. - - - -#### `tf.contrib.learn.RunConfig.__init__(master=None, task=None, num_ps_replicas=None, num_cores=0, log_device_placement=False, gpu_memory_fraction=1, cluster_spec=None, tf_random_seed=None, save_summary_steps=100, save_checkpoints_secs=600, save_checkpoints_steps=None, keep_checkpoint_max=5, keep_checkpoint_every_n_hours=10000, job_name=None, is_chief=None, evaluation_master='')` {#RunConfig.__init__} +#### `tf.contrib.learn.RunConfig.__init__(master=None, num_cores=0, log_device_placement=False, gpu_memory_fraction=1, tf_random_seed=None, save_summary_steps=100, save_checkpoints_secs=600, save_checkpoints_steps=None, keep_checkpoint_max=5, keep_checkpoint_every_n_hours=10000, evaluation_master='')` {#RunConfig.__init__} Constructor. -If set to None, `master`, `task`, `num_ps_replicas`, `cluster_spec`, -`job_name`, and `is_chief` are set based on the TF_CONFIG environment -variable, if the pertinent information is present; otherwise, the defaults -listed in the Args section apply. - -The TF_CONFIG environment variable is a JSON object with two relevant -attributes: `task` and `cluster_spec`. `cluster_spec` is a JSON serialized -version of the Python dict described in server_lib.py. `task` has two -attributes: `type` and `index`, where `type` can be any of the task types -in the cluster_spec. When TF_CONFIG contains said information, the -following properties are set on this class: - - * `job_name` is set to [`task`][`type`] - * `task` is set to [`task`][`index`] - * `cluster_spec` is parsed from [`cluster`] - * 'master' is determined by looking up `job_name` and `task` in the - cluster_spec. - * `num_ps_replicas` is set by counting the number of nodes listed - in the `ps` job of `cluster_spec`. - * `is_chief`: true when `job_name` == "master" and `task` == 0. - -Example: -``` - cluster = {'ps': ['host1:2222', 'host2:2222'], - 'worker': ['host3:2222', 'host4:2222', 'host5:2222']} - os.environ['TF_CONFIG'] = json.dumps({ - {'cluster': cluster, - 'task': {'type': 'worker', 'index': 1}}}) - config = RunConfig() - assert config.master == 'host4:2222' - assert config.task == 1 - assert config.num_ps_replicas == 2 - assert config.cluster_spec == server_lib.ClusterSpec(cluster) - assert config.job_name == 'worker' - assert not config.is_chief -``` +Note that the superclass `ClusterConfig` may set properties like +`cluster_spec`, `is_chief`, `master` (if `None` in the args), +`num_ps_replicas`, `task_id`, and `task_type` based on the `TF_CONFIG` +environment variable. See `ClusterConfig` for more details. ##### Args: * <b>`master`</b>: TensorFlow master. Defaults to empty string for local. -* <b>`task`</b>: Task id of the replica running the training (default: 0). -* <b>`num_ps_replicas`</b>: Number of parameter server tasks to use (default: 0). * <b>`num_cores`</b>: Number of cores to be used. If 0, the system picks an appropriate number (default: 0). * <b>`log_device_placement`</b>: Log the op placement to devices (default: False). * <b>`gpu_memory_fraction`</b>: Fraction of GPU memory used by the process on each GPU uniformly on the same machine. -* <b>`cluster_spec`</b>: a `tf.train.ClusterSpec` object that describes the cluster - in the case of distributed computation. If missing, reasonable - assumptions are made for the addresses of jobs. * <b>`tf_random_seed`</b>: Random seed for TensorFlow initializers. Setting this value allows consistency between reruns. * <b>`save_summary_steps`</b>: Save summaries every this many steps. @@ -74,17 +38,36 @@ Example: * <b>`keep_checkpoint_every_n_hours`</b>: Number of hours between each checkpoint to be saved. The default value of 10,000 hours effectively disables the feature. -* <b>`job_name`</b>: the type of task, e.g., 'ps', 'worker', etc. The `job_name` - must exist in the `cluster_spec.jobs`. -* <b>`is_chief`</b>: whether or not this task (as identified by the other parameters) - should be the chief task. * <b>`evaluation_master`</b>: the master on which to perform evaluation. -##### Raises: + +- - - + +#### `tf.contrib.learn.RunConfig.cluster_spec` {#RunConfig.cluster_spec} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.evaluation_master` {#RunConfig.evaluation_master} -* <b>`ValueError`</b>: if num_ps_replicas and cluster_spec are set (cluster_spec - may come from the TF_CONFIG environment variable). + + +- - - + +#### `tf.contrib.learn.RunConfig.get_task_id()` {#RunConfig.get_task_id} + +Returns task index from `TF_CONFIG` environmental variable. + +If you have a ClusterConfig instance, you can just access its task_id +property instead of calling this function and re-parsing the environmental +variable. + +##### Returns: + + `TF_CONFIG['task']['index']`. Defaults to 0. - - - @@ -96,7 +79,28 @@ Example: - - - -#### `tf.contrib.learn.RunConfig.job_name` {#RunConfig.job_name} +#### `tf.contrib.learn.RunConfig.master` {#RunConfig.master} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.num_ps_replicas` {#RunConfig.num_ps_replicas} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.task_id` {#RunConfig.task_id} + + + + +- - - + +#### `tf.contrib.learn.RunConfig.task_type` {#RunConfig.task_type} diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.decode_base64.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.decode_base64.md index 0d490e313b..1983d467aa 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.decode_base64.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.decode_base64.md @@ -8,10 +8,10 @@ Web-safe means that input must use - and _ instead of + and /. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Base64 strings to decode. +* <b>`input`</b>: An `Output` of type `string`. Base64 strings to decode. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Decoded strings. + An `Output` of type `string`. Decoded strings. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.extract_image_patches.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.extract_image_patches.md index ee24ecd627..bf6f268d4f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.extract_image_patches.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.extract_image_patches.md @@ -5,7 +5,7 @@ Extract `patches` from `images` and put them in the "depth" output dimension. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 4-D Tensor with shape `[batch, in_rows, in_cols, depth]`. * <b>`ksizes`</b>: A list of `ints` that has length `>= 4`. The size of the sliding window for each dimension of `images`. @@ -33,7 +33,7 @@ Extract `patches` from `images` and put them in the "depth" output dimension. ##### Returns: - A `Tensor`. Has the same type as `images`. + A `Output`. Has the same type as `images`. 4-D Tensor with shape `[batch, out_rows, out_cols, ksize_rows * ksize_cols * depth]` containing image patches with size `ksize_rows x ksize_cols x depth` vectorized in the "depth" dimension. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.floor.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.floor.md index 4aadcff6ef..937be96ef3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.floor.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.floor.md @@ -5,10 +5,10 @@ Returns element-wise largest integer not greater than x. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.greater.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.greater.md index 99b34aaca4..eacae8cd6f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.greater.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.greater.md @@ -8,11 +8,11 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.image.hsv_to_rgb.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.image.hsv_to_rgb.md index 9bb9c51198..1803af288c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.image.hsv_to_rgb.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.image.hsv_to_rgb.md @@ -11,11 +11,11 @@ See `rgb_to_hsv` for a description of the HSV encoding. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. 1-D or higher rank. HSV data to convert. Last dimension must be size 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. `images` converted to RGB. + A `Output`. Has the same type as `images`. `images` converted to RGB. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.log.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.log.md index a6c085b5cf..f469657ff4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.log.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.log.md @@ -7,10 +7,10 @@ I.e., \\(y = \log_e x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.depthwise_conv2d_native.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.depthwise_conv2d_native.md index c2736f1ba9..403621dc00 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.depthwise_conv2d_native.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.depthwise_conv2d_native.md @@ -22,8 +22,8 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. * <b>`strides`</b>: A list of `ints`. 1-D of length 4. The stride of the sliding window for each dimension of `input`. @@ -33,5 +33,5 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.dilation2d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.dilation2d.md index b9cf01da19..ef1024d152 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.dilation2d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.dilation2d.md @@ -29,9 +29,9 @@ negation of the erosion of `-input` by the reflected `filter`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 4-D with shape `[batch, in_height, in_width, depth]`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. 3-D with shape `[filter_height, filter_width, depth]`. * <b>`strides`</b>: A list of `ints` that has length `>= 4`. The stride of the sliding window for each dimension of the input @@ -45,6 +45,6 @@ negation of the erosion of `-input` by the reflected `filter`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. 4-D with shape `[batch, out_height, out_width, depth]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.l2_loss.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.l2_loss.md index fd648ca642..626501558a 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.l2_loss.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.l2_loss.md @@ -9,11 +9,11 @@ Computes half the L2 norm of a tensor without the `sqrt`: ##### Args: -* <b>`t`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`t`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Typically 2-D, but may have any dimensions. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `t`. 0-D. + A `Output`. Has the same type as `t`. 0-D. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.max_pool3d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.max_pool3d.md index 960b322c6c..b4ccf50775 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.max_pool3d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.max_pool3d.md @@ -5,7 +5,7 @@ Performs 3D max pooling on the input. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, depth, rows, cols, channels]` tensor to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 5`. 1-D tensor of length 5. The size of the window for each dimension of @@ -19,5 +19,5 @@ Performs 3D max pooling on the input. ##### Returns: - A `Tensor`. Has the same type as `input`. The max pooled output tensor. + A `Output`. Has the same type as `input`. The max pooled output tensor. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.softplus.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.softplus.md index c0faef9687..0e8db897c6 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.softplus.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.softplus.md @@ -5,10 +5,10 @@ Computes softplus: `log(exp(features) + 1)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.placeholder_with_default.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.placeholder_with_default.md index 2719b876f1..5b8cf54457 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.placeholder_with_default.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.placeholder_with_default.md @@ -5,13 +5,13 @@ A placeholder op that passes though `input` when its output is not fed. ##### Args: -* <b>`input`</b>: A `Tensor`. The default value to produce when `output` is not fed. +* <b>`input`</b>: A `Output`. The default value to produce when `output` is not fed. * <b>`shape`</b>: A `tf.TensorShape` or list of `ints`. The (possibly partial) shape of the tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. A placeholder tensor that defaults to `input` if it is not fed. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.segment_max.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.segment_max.md index c9d7a28900..1bd395b6c5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.segment_max.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.segment_max.md @@ -16,15 +16,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.select.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.select.md index af9d8dbb76..78d1923e21 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.select.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.select.md @@ -45,14 +45,14 @@ select(condition, t, e) ==> [[1, 2], ##### Args: -* <b>`condition`</b>: A `Tensor` of type `bool`. -* <b>`t`</b>: A `Tensor` which may have the same shape as `condition`. +* <b>`condition`</b>: An `Output` of type `bool`. +* <b>`t`</b>: An `Output` which may have the same shape as `condition`. If `condition` is rank 1, `t` may have higher rank, but its first dimension must match the size of `condition`. -* <b>`e`</b>: A `Tensor` with the same type and shape as `t`. +* <b>`e`</b>: An `Output` with the same type and shape as `t`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` with the same type and shape as `t` and `e`. + An `Output` with the same type and shape as `t` and `e`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.string_to_hash_bucket_fast.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.string_to_hash_bucket_fast.md index e684058326..58436df212 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.string_to_hash_bucket_fast.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.string_to_hash_bucket_fast.md @@ -12,12 +12,12 @@ to the same bucket. To prevent this problem, use a strong hash function with ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. The strings to assign a hash bucket. +* <b>`input`</b>: An `Output` of type `string`. The strings to assign a hash bucket. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.sub.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.sub.md index 83dbd7a93c..14503d9682 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.sub.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.sub.md @@ -8,11 +8,11 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.tile.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.tile.md index 0c31e73c98..a7c01939e2 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.tile.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.tile.md @@ -11,12 +11,12 @@ dimension. For example, tiling `[a b c d]` by `[2]` produces ##### Args: -* <b>`input`</b>: A `Tensor`. 1-D or higher. -* <b>`multiples`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. 1-D or higher. +* <b>`multiples`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. Length must be the same as the number of dimensions in `input` * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.unsorted_segment_sum.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.unsorted_segment_sum.md index c02d39e96a..781a4ff1fb 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.unsorted_segment_sum.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.unsorted_segment_sum.md @@ -23,15 +23,15 @@ If the sum is empty for a given segment ID `i`, `output[i] = 0`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor whose shape is a prefix of `data.shape`. -* <b>`num_segments`</b>: A `Tensor` of type `int32`. +* <b>`num_segments`</b>: An `Output` of type `int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for the first `segment_ids.rank` dimensions, which are replaced with a single dimension which has size `num_segments`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.add.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.add.md index da82da6076..d0ee8c4783 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.add.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.add.md @@ -8,11 +8,11 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.asin.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.asin.md index 64ec024b4c..d23d7077fc 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.asin.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.asin.md @@ -5,10 +5,10 @@ Computes asin of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.batch_matmul.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.batch_matmul.md index a4764435b8..316ef3f58e 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.batch_matmul.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.batch_matmul.md @@ -2,7 +2,7 @@ Multiplies slices of two tensors in batches. -Multiplies all slices of `Tensor` `x` and `y` (each slice can be +Multiplies all slices of `Output` `x` and `y` (each slice can be viewed as an element of a batch), and arranges the individual results in a single output tensor of the same batch size. Each of the individual slices can optionally be adjointed (to adjoint a matrix @@ -24,9 +24,9 @@ It is computed as: ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `complex64`, `complex128`. 3-D or higher with shape `[..., r_x, c_x]`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. 3-D or higher with shape `[..., r_y, c_y]`. * <b>`adj_x`</b>: An optional `bool`. Defaults to `False`. If `True`, adjoint the slices of `x`. Defaults to `False`. @@ -36,6 +36,6 @@ It is computed as: ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. 3-D or higher with shape `[..., r_o, c_o]` diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.cos.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.cos.md index faf84ea9d3..3787d628f9 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.cos.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.cos.md @@ -5,10 +5,10 @@ Computes cos of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.diag_part.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.diag_part.md index 845a45669b..8e1acfe99c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.diag_part.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.diag_part.md @@ -24,11 +24,11 @@ tf.diag_part(input) ==> [1, 2, 3, 4] ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. Rank k tensor where k is 2, 4, or 6. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. The extracted diagonal. + A `Output`. Has the same type as `input`. The extracted diagonal. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.div.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.div.md index bc34738d56..56b9cc1aa0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.div.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.div.md @@ -8,11 +8,11 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.resize_nearest_neighbor.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.resize_nearest_neighbor.md index ba72e73ebd..9e89aab5f1 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.resize_nearest_neighbor.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.resize_nearest_neighbor.md @@ -5,7 +5,7 @@ Resize `images` to `size` using nearest neighbor interpolation. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -17,6 +17,6 @@ Resize `images` to `size` using nearest neighbor interpolation. ##### Returns: - A `Tensor`. Has the same type as `images`. 4-D with shape + A `Output`. Has the same type as `images`. 4-D with shape `[batch, new_height, new_width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.sample_distorted_bounding_box.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.sample_distorted_bounding_box.md index 96c0e65950..2b6f8aa531 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.sample_distorted_bounding_box.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.sample_distorted_bounding_box.md @@ -44,9 +44,9 @@ false and no bounding boxes are supplied, an error is raised. ##### Args: -* <b>`image_size`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`. +* <b>`image_size`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`. 1-D, containing `[height, width, channels]`. -* <b>`bounding_boxes`</b>: A `Tensor` of type `float32`. +* <b>`bounding_boxes`</b>: An `Output` of type `float32`. 3-D with shape `[batch, N, 4]` describing the N bounding boxes associated with the image. * <b>`seed`</b>: An optional `int`. Defaults to `0`. @@ -76,12 +76,12 @@ false and no bounding boxes are supplied, an error is raised. ##### Returns: - A tuple of `Tensor` objects (begin, size, bboxes). + A tuple of `Output` objects (begin, size, bboxes). -* <b>`begin`</b>: A `Tensor`. Has the same type as `image_size`. 1-D, containing `[offset_height, offset_width, 0]`. Provide as input to +* <b>`begin`</b>: A `Output`. Has the same type as `image_size`. 1-D, containing `[offset_height, offset_width, 0]`. Provide as input to `tf.slice`. -* <b>`size`</b>: A `Tensor`. Has the same type as `image_size`. 1-D, containing `[target_height, target_width, -1]`. Provide as input to +* <b>`size`</b>: A `Output`. Has the same type as `image_size`. 1-D, containing `[target_height, target_width, -1]`. Provide as input to `tf.slice`. -* <b>`bboxes`</b>: A `Tensor` of type `float32`. 3-D with shape `[1, 1, 4]` containing the distorted bounding box. +* <b>`bboxes`</b>: An `Output` of type `float32`. 3-D with shape `[1, 1, 4]` containing the distorted bounding box. Provide as input to `tf.image.draw_bounding_boxes`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_and.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_and.md index 2b5f011ccd..67dad9d598 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_and.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_and.md @@ -8,11 +8,11 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_not.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_not.md index 40a0bb2e43..f624b693c4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_not.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_not.md @@ -5,10 +5,10 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.nn.conv3d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.nn.conv3d.md index cbac47eb58..79ddefed06 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.nn.conv3d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.nn.conv3d.md @@ -11,9 +11,9 @@ Our Conv3D implements a form of cross-correlation. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, in_depth, in_height, in_width, in_channels]`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. Shape `[filter_depth, filter_height, filter_width, in_channels, out_channels]`. `in_channels` must match between `input` and `filter`. * <b>`strides`</b>: A list of `ints` that has length `>= 5`. @@ -25,5 +25,5 @@ Our Conv3D implements a form of cross-correlation. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.reverse_v2.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.reverse_v2.md index 48d0e68e20..218d059e7f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.reverse_v2.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.reverse_v2.md @@ -49,13 +49,13 @@ reverse(t, dims) ==> [[[[8, 9, 10, 11], ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. Up to 8-D. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. The indices of the dimensions to reverse. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. The same shape as `tensor`. + A `Output`. Has the same type as `tensor`. The same shape as `tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.scatter_nd.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.scatter_nd.md index c4d448d9d8..fac7357e02 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.scatter_nd.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.scatter_nd.md @@ -9,7 +9,7 @@ operator which extracts values or slices from a given tensor. TODO(simister): Add a link to Variable.__getitem__ documentation on slice syntax. -`shape` is a `TensorShape` with rank `P` and `indices` is a `Tensor` of rank +`shape` is a `TensorShape` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `shape`. @@ -76,19 +76,19 @@ The resulting tensor would look like this: ##### Args: -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. +* <b>`updates`</b>: A `Output`. A Tensor. Must have the same type as tensor. A tensor of updated values to store in ref. -* <b>`shape`</b>: A `Tensor`. Must have the same type as `indices`. +* <b>`shape`</b>: A `Output`. Must have the same type as `indices`. A vector. The shape of the resulting tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `updates`. + A `Output`. Has the same type as `updates`. A new tensor with the given shape and updates applied according to the indices. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.tan.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.tan.md index cb05f1427b..6bc078c0c3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.tan.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.tan.md @@ -5,10 +5,10 @@ Computes tan of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.abs.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.abs.md index 58c7d0521f..8d6b780f1c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.abs.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.abs.md @@ -7,7 +7,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.as_string.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.as_string.md index 0217ad3113..289ab255e0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.as_string.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.as_string.md @@ -7,7 +7,7 @@ types and boolean. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `complex64`, `float32`, `float64`, `bool`, `int8`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `complex64`, `float32`, `float64`, `bool`, `int8`. * <b>`precision`</b>: An optional `int`. Defaults to `-1`. The post-decimal precision to use for floating point numbers. Only used if precision > -1. @@ -27,5 +27,5 @@ types and boolean. ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.bitcast.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.bitcast.md index 9e60ab2144..1c989fd6d8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.bitcast.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.bitcast.md @@ -18,11 +18,11 @@ endian orderings will give different results. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. * <b>`type`</b>: A `tf.DType` from: `tf.float32, tf.float64, tf.int64, tf.int32, tf.uint8, tf.uint16, tf.int16, tf.int8, tf.complex64, tf.complex128, tf.qint8, tf.quint8, tf.qint32, tf.half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `type`. + A `Output` of type `type`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.learn.RunConfig.get_task_id.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.learn.RunConfig.get_task_id.md new file mode 100644 index 0000000000..1c2856df21 --- /dev/null +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.learn.RunConfig.get_task_id.md @@ -0,0 +1,12 @@ +#### `tf.contrib.learn.RunConfig.get_task_id()` {#RunConfig.get_task_id} + +Returns task index from `TF_CONFIG` environmental variable. + +If you have a ClusterConfig instance, you can just access its task_id +property instead of calling this function and re-parsing the environmental +variable. + +##### Returns: + + `TF_CONFIG['task']['index']`. Defaults to 0. + diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.decode_json_example.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.decode_json_example.md index bf5184c40a..dd6a7a067b 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.decode_json_example.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.decode_json_example.md @@ -12,14 +12,14 @@ Example-parsing ops. ##### Args: -* <b>`json_examples`</b>: A `Tensor` of type `string`. +* <b>`json_examples`</b>: An `Output` of type `string`. Each string is a JSON object serialized according to the JSON mapping of the Example proto. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. Each string is a binary Example protocol buffer corresponding to the respective element of `json_examples`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dequantize.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dequantize.md index edf0de7a04..2b0274e2a9 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dequantize.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dequantize.md @@ -38,15 +38,15 @@ result = range_min + ((input - numeric_limits<T>::min()) * range_scale) ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. -* <b>`min_range`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`min_range`</b>: An `Output` of type `float32`. The minimum scalar value possibly produced for the input. -* <b>`max_range`</b>: A `Tensor` of type `float32`. +* <b>`max_range`</b>: An `Output` of type `float32`. The maximum scalar value possibly produced for the input. * <b>`mode`</b>: An optional `string` from: `"MIN_COMBINED", "MIN_FIRST"`. Defaults to `"MIN_COMBINED"`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dynamic_partition.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dynamic_partition.md index e24bc8c39e..1233164663 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dynamic_partition.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.dynamic_partition.md @@ -41,8 +41,8 @@ For example: ##### Args: -* <b>`data`</b>: A `Tensor`. -* <b>`partitions`</b>: A `Tensor` of type `int32`. +* <b>`data`</b>: A `Output`. +* <b>`partitions`</b>: An `Output` of type `int32`. Any shape. Indices in the range `[0, num_partitions)`. * <b>`num_partitions`</b>: An `int` that is `>= 1`. The number of partitions to output. @@ -50,5 +50,5 @@ For example: ##### Returns: - A list of `num_partitions` `Tensor` objects of the same type as data. + A list of `num_partitions` `Output` objects of the same type as data. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.erfc.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.erfc.md index 62c13418f7..47b4f44ee0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.erfc.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.erfc.md @@ -5,10 +5,10 @@ Computes the complementary error function of `x` element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.fake_quant_with_min_max_args.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.fake_quant_with_min_max_args.md index fcad8cb500..2d19701ac4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.fake_quant_with_min_max_args.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.fake_quant_with_min_max_args.md @@ -11,12 +11,12 @@ Quantization is called fake since the output is still in floating point. ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `float32`. +* <b>`inputs`</b>: An `Output` of type `float32`. * <b>`min`</b>: An optional `float`. Defaults to `-6`. * <b>`max`</b>: An optional `float`. Defaults to `6`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.igamma.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.igamma.md index 92b5fbe851..601b90a3f5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.igamma.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.igamma.md @@ -19,11 +19,11 @@ Gamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.decode_gif.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.decode_gif.md index 45e7ab9d22..244683b3cc 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.decode_gif.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.decode_gif.md @@ -10,11 +10,11 @@ convert $src.gif -coalesce $dst.gif ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The GIF-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The GIF-encoded image. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `uint8`. + An `Output` of type `uint8`. 4-D with shape `[num_frames, height, width, 3]`. RGB order diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.extract_glimpse.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.extract_glimpse.md index 83482124e7..77da7e295c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.extract_glimpse.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.extract_glimpse.md @@ -27,13 +27,13 @@ The argument `normalized` and `centered` controls how the windows are built: ##### Args: -* <b>`input`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: An `Output` of type `float32`. A 4-D float tensor of shape `[batch_size, height, width, channels]`. -* <b>`size`</b>: A `Tensor` of type `int32`. +* <b>`size`</b>: An `Output` of type `int32`. A 1-D tensor of 2 elements containing the size of the glimpses to extract. The glimpse height must be specified first, following by the glimpse width. -* <b>`offsets`</b>: A `Tensor` of type `float32`. +* <b>`offsets`</b>: An `Output` of type `float32`. A 2-D integer tensor of shape `[batch_size, 2]` containing the x, y locations of the center of each window. * <b>`centered`</b>: An optional `bool`. Defaults to `True`. @@ -50,7 +50,7 @@ The argument `normalized` and `centered` controls how the windows are built: ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. A tensor representing the glimpses `[batch_size, glimpse_height, glimpse_width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.rgb_to_hsv.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.rgb_to_hsv.md index c08a086b88..1009faea54 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.rgb_to_hsv.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.image.rgb_to_hsv.md @@ -13,11 +13,11 @@ corresponds to pure red, hue 1/3 is pure green, and 2/3 is pure blue. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. 1-D or higher rank. RGB data to convert. Last dimension must be size 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. `images` converted to HSV. + A `Output`. Has the same type as `images`. `images` converted to HSV. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.maximum.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.maximum.md index aec816dcba..89fe27e2de 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.maximum.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.maximum.md @@ -8,11 +8,11 @@ Returns the max of x and y (i.e. x > y ? x : y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.multiply.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.multiply.md index aea6264f04..0d34d34ea6 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.multiply.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.multiply.md @@ -8,11 +8,11 @@ Returns x * y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.elu.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.elu.md index 8ffeeca65c..bfeaade123 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.elu.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.elu.md @@ -8,10 +8,10 @@ See [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs) ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.reciprocal.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.reciprocal.md index d340aa5178..210e4aa6d8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.reciprocal.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.reciprocal.md @@ -7,10 +7,10 @@ I.e., \\(y = 1 / x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.space_to_depth.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.space_to_depth.md index afffa0d148..fc6d3df93c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.space_to_depth.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.space_to_depth.md @@ -77,11 +77,11 @@ x = [[[[1, 2, 3, 4], ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`block_size`</b>: An `int` that is `>= 2`. The size of the spatial block. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_mean.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_mean.md index af7affaa9f..ad8aeaae32 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_mean.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_mean.md @@ -12,16 +12,16 @@ dimension, selecting a subset of dimension 0, specified by `indices`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_sum.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_sum.md index e48ae891c3..a664300c61 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_sum.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.sparse_segment_sum.md @@ -35,16 +35,16 @@ tf.segment_sum(c, tf.constant([0, 0, 1])) ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.string_join.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.string_join.md index b81537a70c..a46cce03a0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.string_join.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.string_join.md @@ -7,7 +7,7 @@ with the given separator (default is an empty separator). ##### Args: -* <b>`inputs`</b>: A list of at least 1 `Tensor` objects of type `string`. +* <b>`inputs`</b>: A list of at least 1 `Output` objects of type `string`. A list of string tensors. The tensors must all have the same shape, or be scalars. Scalars may be mixed in; these will be broadcast to the shape of non-scalar inputs. @@ -17,5 +17,5 @@ with the given separator (default is an empty separator). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.Output.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.Output.md index 29ace7b396..144733b975 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.Output.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.Output.md @@ -210,7 +210,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -238,13 +239,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -259,13 +260,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -306,13 +307,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -369,13 +370,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -450,13 +451,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -495,12 +496,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -532,13 +533,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -553,13 +554,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -574,13 +575,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -601,12 +602,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -634,13 +635,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -684,13 +685,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -705,13 +706,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -726,13 +727,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -789,13 +790,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -817,13 +818,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -867,13 +868,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -936,13 +937,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.count_up_to.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.count_up_to.md index 97f802372c..527170c8ef 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.count_up_to.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.count_up_to.md @@ -5,7 +5,7 @@ Increments 'ref' until it reaches 'limit'. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `int32`, `int64`. Should be from a scalar `Variable` node. * <b>`limit`</b>: An `int`. If incrementing ref would bring it above limit, instead generates an @@ -14,7 +14,7 @@ Increments 'ref' until it reaches 'limit'. ##### Returns: - A `Tensor`. Has the same type as `ref`. + A `Output`. Has the same type as `ref`. A copy of the input before increment. If nothing else modifies the input, the values produced will all be distinct. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.dynamic_stitch.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.dynamic_stitch.md index 3eaba84d7c..d199c7a145 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.dynamic_stitch.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.dynamic_stitch.md @@ -49,11 +49,11 @@ For example: ##### Args: -* <b>`indices`</b>: A list of at least 1 `Tensor` objects of type `int32`. -* <b>`data`</b>: A list with the same number of `Tensor` objects as `indices` of `Tensor` objects of the same type. +* <b>`indices`</b>: A list of at least 1 `Output` objects of type `int32`. +* <b>`data`</b>: A list with the same number of `Output` objects as `indices` of `Output` objects of the same type. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.fft3d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.fft3d.md index a1cf358fe2..542867e26f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.fft3d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.fft3d.md @@ -7,12 +7,12 @@ dimensions of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 3 dimensions of `input` are replaced with their 3D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.ifft.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.ifft.md index 4e8b5c691d..7df1ce028c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.ifft.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.ifft.md @@ -7,12 +7,12 @@ dimension of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most dimension of `input` is replaced with its inverse 1D Fourier Transform. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.non_max_suppression.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.non_max_suppression.md index d6b354a3d2..2ab6ce96db 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.non_max_suppression.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.non_max_suppression.md @@ -24,12 +24,12 @@ using the `tf.gather operation`. For example: ##### Args: -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. A 2-D float tensor of shape `[num_boxes, 4]`. -* <b>`scores`</b>: A `Tensor` of type `float32`. +* <b>`scores`</b>: An `Output` of type `float32`. A 1-D float tensor of shape `[num_boxes]` representing a single score corresponding to each box (each row of boxes). -* <b>`max_output_size`</b>: A `Tensor` of type `int32`. +* <b>`max_output_size`</b>: An `Output` of type `int32`. A scalar integer tensor representing the maximum number of boxes to be selected by non max suppression. * <b>`iou_threshold`</b>: An optional `float`. Defaults to `0.5`. @@ -39,7 +39,7 @@ using the `tf.gather operation`. For example: ##### Returns: - A `Tensor` of type `int32`. + An `Output` of type `int32`. A 1-D integer tensor of shape `[M]` representing the selected indices from the boxes tensor, where `M <= max_output_size`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.resize_area.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.resize_area.md index dbc6fd1bcd..bccb87bcb3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.resize_area.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.image.resize_area.md @@ -7,7 +7,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -19,6 +19,6 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.less.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.less.md index 3a00afa8db..39b804ecce 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.less.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.less.md @@ -8,11 +8,11 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.lgamma.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.lgamma.md index a4add48fb4..60f5b4b571 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.lgamma.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.lgamma.md @@ -5,10 +5,10 @@ Computes the log of the absolute value of `Gamma(x)` element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.logical_or.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.logical_or.md index e04b6a15d2..8ef735af15 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.logical_or.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.logical_or.md @@ -8,11 +8,11 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_band_part.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_band_part.md index 87bd745c2a..2770341a6f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_band_part.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_band_part.md @@ -45,17 +45,17 @@ Useful special cases: ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k` tensor. -* <b>`num_lower`</b>: A `Tensor` of type `int64`. +* <b>`input`</b>: A `Output`. Rank `k` tensor. +* <b>`num_lower`</b>: An `Output` of type `int64`. 0-D tensor. Number of subdiagonals to keep. If negative, keep entire lower triangle. -* <b>`num_upper`</b>: A `Tensor` of type `int64`. +* <b>`num_upper`</b>: An `Output` of type `int64`. 0-D tensor. Number of superdiagonals to keep. If negative, keep entire upper triangle. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Rank `k` tensor of the same shape as input. The extracted banded tensor. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag.md index 16ba620c83..c72b7a2c35 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag.md @@ -32,11 +32,11 @@ which has shape (2, 4, 4) ##### Args: -* <b>`diagonal`</b>: A `Tensor`. Rank `k`, where `k >= 1`. +* <b>`diagonal`</b>: A `Output`. Rank `k`, where `k >= 1`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `diagonal`. + A `Output`. Has the same type as `diagonal`. Rank `k+1`, with `output.shape = diagonal.shape + [diagonal.shape[-1]]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag_part.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag_part.md index efaf772f6b..f91c560c12 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag_part.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_diag_part.md @@ -34,12 +34,12 @@ which has shape (2, 4) ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k` tensor where `k >= 2`. +* <b>`input`</b>: A `Output`. Rank `k` tensor where `k >= 2`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The extracted diagonal(s) having shape `diagonal.shape = input.shape[:-2] + [min(input.shape[-2:])]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_solve.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_solve.md index 88d037f2fa..439ca2d065 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_solve.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_solve.md @@ -12,9 +12,9 @@ If `adjoint` is `True` then each output matrix satisfies ##### Args: -* <b>`matrix`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`, `complex64`, `complex128`. +* <b>`matrix`</b>: A `Output`. Must be one of the following types: `float64`, `float32`, `complex64`, `complex128`. Shape is `[..., M, M]`. -* <b>`rhs`</b>: A `Tensor`. Must have the same type as `matrix`. +* <b>`rhs`</b>: A `Output`. Must have the same type as `matrix`. Shape is `[..., M, K]`. * <b>`adjoint`</b>: An optional `bool`. Defaults to `False`. Boolean indicating whether to solve with `matrix` or its (block-wise) @@ -23,5 +23,5 @@ If `adjoint` is `True` then each output matrix satisfies ##### Returns: - A `Tensor`. Has the same type as `matrix`. Shape is `[..., M, K]`. + A `Output`. Has the same type as `matrix`. Shape is `[..., M, K]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.fractional_avg_pool.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.fractional_avg_pool.md index 367205ffd6..5c1a78059b 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.fractional_avg_pool.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.fractional_avg_pool.md @@ -10,7 +10,7 @@ pooling region. ##### Args: -* <b>`value`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. +* <b>`value`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. 4-D with shape `[batch, height, width, channels]`. * <b>`pooling_ratio`</b>: A list of `floats` that has length `>= 4`. Pooling ratio for each dimension of `value`, currently only @@ -49,9 +49,9 @@ pooling region. ##### Returns: - A tuple of `Tensor` objects (output, row_pooling_sequence, col_pooling_sequence). + A tuple of `Output` objects (output, row_pooling_sequence, col_pooling_sequence). -* <b>`output`</b>: A `Tensor`. Has the same type as `value`. output tensor after fractional avg pooling. -* <b>`row_pooling_sequence`</b>: A `Tensor` of type `int64`. row pooling sequence, needed to calculate gradient. -* <b>`col_pooling_sequence`</b>: A `Tensor` of type `int64`. column pooling sequence, needed to calculate gradient. +* <b>`output`</b>: A `Output`. Has the same type as `value`. output tensor after fractional avg pooling. +* <b>`row_pooling_sequence`</b>: An `Output` of type `int64`. row pooling sequence, needed to calculate gradient. +* <b>`col_pooling_sequence`</b>: An `Output` of type `int64`. column pooling sequence, needed to calculate gradient. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.in_top_k.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.in_top_k.md index f46780649d..88a8087873 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.in_top_k.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.in_top_k.md @@ -20,14 +20,14 @@ $$out_i = predictions_{i, targets_i} \in TopKIncludingTies(predictions_i)$$ ##### Args: -* <b>`predictions`</b>: A `Tensor` of type `float32`. +* <b>`predictions`</b>: An `Output` of type `float32`. A `batch_size` x `classes` tensor. -* <b>`targets`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`targets`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A `batch_size` vector of class ids. * <b>`k`</b>: An `int`. Number of top elements to look at for computing precision. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. Computed Precision at `k` as a `bool Tensor`. + An `Output` of type `bool`. Computed Precision at `k` as a `bool Tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.local_response_normalization.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.local_response_normalization.md index 81134df29f..4feb46302d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.local_response_normalization.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.local_response_normalization.md @@ -17,7 +17,7 @@ convolutional neural networks (NIPS 2012)](http://papers.nips.cc/paper/4824-imag ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D. * <b>`depth_radius`</b>: An optional `int`. Defaults to `5`. 0-D. Half-width of the 1-D normalization window. @@ -30,5 +30,5 @@ convolutional neural networks (NIPS 2012)](http://papers.nips.cc/paper/4824-imag ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.quantized_avg_pool.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.quantized_avg_pool.md index 4bc6a1dc6d..6c6fc38a2b 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.quantized_avg_pool.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.nn.quantized_avg_pool.md @@ -5,11 +5,11 @@ Produces the average pool of the input tensor for quantized types. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. 4-D with shape `[batch, height, width, channels]`. -* <b>`min_input`</b>: A `Tensor` of type `float32`. +* <b>`min_input`</b>: An `Output` of type `float32`. The float value that the lowest quantized input value represents. -* <b>`max_input`</b>: A `Tensor` of type `float32`. +* <b>`max_input`</b>: An `Output` of type `float32`. The float value that the highest quantized input value represents. * <b>`ksize`</b>: A list of `ints`. The size of the window for each dimension of the input tensor. @@ -23,9 +23,9 @@ Produces the average pool of the input tensor for quantized types. ##### Returns: - A tuple of `Tensor` objects (output, min_output, max_output). + A tuple of `Output` objects (output, min_output, max_output). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. -* <b>`min_output`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized output value represents. -* <b>`max_output`</b>: A `Tensor` of type `float32`. The float value that the highest quantized output value represents. +* <b>`output`</b>: A `Output`. Has the same type as `input`. +* <b>`min_output`</b>: An `Output` of type `float32`. The float value that the lowest quantized output value represents. +* <b>`max_output`</b>: An `Output` of type `float32`. The float value that the highest quantized output value represents. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.polygamma.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.polygamma.md index c8b5b2578a..161e7fe2c3 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.polygamma.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.polygamma.md @@ -12,11 +12,11 @@ where \\(\psi(x)\\) is the digamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_nd_update.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_nd_update.md index e7e975cbd3..b6dbcdf997 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_nd_update.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_nd_update.md @@ -4,7 +4,7 @@ Applies sparse `updates` to individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -13,7 +13,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -39,11 +39,11 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. A mutable Tensor. Should be from a Variable node. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to add to ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `True`. @@ -54,7 +54,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_sub.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_sub.md index 8f1afc42f6..bb0b0d9b3f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_sub.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.scatter_sub.md @@ -26,11 +26,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to subtract from `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the subtraction will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_prod.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_prod.md index c1e3e74cf5..a0c80ca18e 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_prod.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_prod.md @@ -17,15 +17,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_sum.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_sum.md index be93c31a2e..c1c83baa3c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_sum.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.segment_sum.md @@ -16,15 +16,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.string_to_hash_bucket.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.string_to_hash_bucket.md index 1b818c2d3b..eb3aa2f97c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.string_to_hash_bucket.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.string_to_hash_bucket.md @@ -12,12 +12,12 @@ This functionality will be deprecated and it's recommended to use ##### Args: -* <b>`string_tensor`</b>: A `Tensor` of type `string`. +* <b>`string_tensor`</b>: An `Output` of type `string`. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.unique.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.unique.md index 5b9bc642c8..4cba0787b2 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.unique.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard7/tf.unique.md @@ -21,14 +21,14 @@ idx ==> [0, 0, 1, 2, 2, 2, 3, 4, 4] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. +* <b>`x`</b>: A `Output`. 1-D. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (y, idx). + A tuple of `Output` objects (y, idx). -* <b>`y`</b>: A `Tensor`. Has the same type as `x`. 1-D. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. +* <b>`y`</b>: A `Output`. Has the same type as `x`. 1-D. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.Variable.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.Variable.md index df09adb501..c62af2f41f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.Variable.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.Variable.md @@ -392,7 +392,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -420,13 +421,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -441,13 +442,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -462,13 +463,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -518,13 +519,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -590,13 +591,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -608,12 +609,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -644,13 +645,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -665,13 +666,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -686,13 +687,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -713,12 +714,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -733,13 +734,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -783,13 +784,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -804,13 +805,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -825,13 +826,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -881,13 +882,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -909,13 +910,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -959,13 +960,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1028,13 +1029,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.acos.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.acos.md index 15ecc97044..166c1bb784 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.acos.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.acos.md @@ -5,10 +5,10 @@ Computes acos of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.argmax.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.argmax.md index 8e92fb59e8..7eb2c2be27 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.argmax.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.argmax.md @@ -5,13 +5,13 @@ Returns the index with the largest value across axiss of a tensor. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. int32, 0 <= axis < rank(input). Describes which axis of the input Tensor to reduce across. For vectors, use axis = 0. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.assign_sub.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.assign_sub.md index b24da4db87..e891c589f1 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.assign_sub.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.assign_sub.md @@ -8,9 +8,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be subtracted to the variable. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the subtraction will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.atan.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.atan.md index 63fe76f460..6c9a41cdd6 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.atan.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.atan.md @@ -5,10 +5,10 @@ Computes atan of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.batch_to_space_nd.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.batch_to_space_nd.md index 40f53ba775..a5131a6e10 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.batch_to_space_nd.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.batch_to_space_nd.md @@ -12,12 +12,12 @@ reverse of SpaceToBatch. See below for a precise description. ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`, where spatial_shape has M dimensions. -* <b>`block_shape`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`block_shape`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D with shape `[M]`, all values must be >= 1. -* <b>`crops`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`crops`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D with shape `[M, 2]`, all values must be >= 0. `crops[i] = [crop_start, crop_end]` specifies the amount to crop from input dimension `i + 1`, which corresponds to spatial dimension `i`. It is @@ -132,5 +132,5 @@ reverse of SpaceToBatch. See below for a precise description. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.cross.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.cross.md index eecf2e869b..923faa253d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.cross.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.cross.md @@ -9,14 +9,14 @@ of corresponding 3-element vectors is cross-multiplied independently. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. A tensor containing 3-element vectors. -* <b>`b`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`b`</b>: A `Output`. Must have the same type as `a`. Another tensor, of same type and shape as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. Pairwise cross product of the vectors in `a` and `b`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.equal.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.equal.md index 332a12f725..2073a80e91 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.equal.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.equal.md @@ -8,11 +8,11 @@ Returns the truth value of (x == y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.draw_bounding_boxes.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.draw_bounding_boxes.md index fff67cd42f..37de28a955 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.draw_bounding_boxes.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.draw_bounding_boxes.md @@ -17,16 +17,16 @@ Parts of the bounding box may fall outside the image. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D with shape `[batch, height, width, depth]`. A batch of images. -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. 3-D with shape `[batch, num_bounding_boxes, 4]` containing bounding boxes. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. + A `Output`. Has the same type as `images`. 4-D with the same shape as `images`. The batch of input images with bounding boxes drawn on the images. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.resize_bilinear.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.resize_bilinear.md index a9580ca199..93b11fc83a 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.resize_bilinear.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.image.resize_bilinear.md @@ -7,7 +7,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -19,6 +19,6 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.is_inf.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.is_inf.md index 56663d6417..37ed3bbf8a 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.is_inf.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.is_inf.md @@ -9,10 +9,10 @@ Equivalent to np.isinf ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.less_equal.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.less_equal.md index c8ce84b669..b102b1d717 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.less_equal.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.less_equal.md @@ -8,11 +8,11 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_inverse.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_inverse.md index ff49493f0c..31b295bc7e 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_inverse.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_inverse.md @@ -17,14 +17,14 @@ garbage result. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. * <b>`adjoint`</b>: An optional `bool`. Defaults to `False`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[..., M, M]`. + A `Output`. Has the same type as `input`. Shape is `[..., M, M]`. @compatibility(numpy) Equivalent to np.linalg.inv diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_set_diag.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_set_diag.md index a8f9bf6be8..14b7b6524c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_set_diag.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.matrix_set_diag.md @@ -18,13 +18,13 @@ tensor of rank `k+1` with dimensions `[I, J, K, ..., M, N]` where: ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k+1`, where `k >= 1`. -* <b>`diagonal`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Rank `k+1`, where `k >= 1`. +* <b>`diagonal`</b>: A `Output`. Must have the same type as `input`. Rank `k`, where `k >= 1`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Rank `k+1`, with `output.shape = input.shape`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.conv2d.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.conv2d.md index d40ed35657..a7d2cf094f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.conv2d.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.conv2d.md @@ -27,8 +27,8 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. * <b>`strides`</b>: A list of `ints`. 1-D of length 4. The stride of the sliding window for each dimension of `input`. Must be in the same order as the dimension specified with format. @@ -45,5 +45,5 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.quantized_concat.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.quantized_concat.md index 0bb94d727d..582c6589d4 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.quantized_concat.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.quantized_concat.md @@ -5,25 +5,25 @@ Concatenates quantized tensors along one dimension. ##### Args: -* <b>`concat_dim`</b>: A `Tensor` of type `int32`. +* <b>`concat_dim`</b>: An `Output` of type `int32`. 0-D. The dimension along which to concatenate. Must be in the range [0, rank(values)). -* <b>`values`</b>: A list of at least 2 `Tensor` objects of the same type. +* <b>`values`</b>: A list of at least 2 `Output` objects of the same type. The `N` Tensors to concatenate. Their ranks and types must match, and their sizes must match in all dimensions except `concat_dim`. -* <b>`input_mins`</b>: A list with the same number of `Tensor` objects as `values` of `Tensor` objects of type `float32`. +* <b>`input_mins`</b>: A list with the same number of `Output` objects as `values` of `Output` objects of type `float32`. The minimum scalar values for each of the input tensors. -* <b>`input_maxes`</b>: A list with the same number of `Tensor` objects as `values` of `Tensor` objects of type `float32`. +* <b>`input_maxes`</b>: A list with the same number of `Output` objects as `values` of `Output` objects of type `float32`. The maximum scalar values for each of the input tensors. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (output, output_min, output_max). + A tuple of `Output` objects (output, output_min, output_max). -* <b>`output`</b>: A `Tensor`. Has the same type as `values`. A `Tensor` with the concatenation of values stacked along the +* <b>`output`</b>: A `Output`. Has the same type as `values`. An `Output` with the concatenation of values stacked along the `concat_dim` dimension. This tensor's shape matches that of `values` except in `concat_dim` where it has the sum of the sizes. -* <b>`output_min`</b>: A `Tensor` of type `float32`. The float value that the minimum quantized output value represents. -* <b>`output_max`</b>: A `Tensor` of type `float32`. The float value that the maximum quantized output value represents. +* <b>`output_min`</b>: An `Output` of type `float32`. The float value that the minimum quantized output value represents. +* <b>`output_max`</b>: An `Output` of type `float32`. The float value that the maximum quantized output value represents. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.reverse.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.reverse.md index f3dfd41b31..f139d3406f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.reverse.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.reverse.md @@ -50,12 +50,12 @@ reverse(t, dims) ==> [[[[8, 9, 10, 11], ##### Args: -* <b>`tensor`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`tensor`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int32`, `int64`, `bool`, `half`, `float32`, `float64`, `complex64`, `complex128`. Up to 8-D. -* <b>`dims`</b>: A `Tensor` of type `bool`. 1-D. The dimensions to reverse. +* <b>`dims`</b>: An `Output` of type `bool`. 1-D. The dimensions to reverse. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `tensor`. The same shape as `tensor`. + A `Output`. Has the same type as `tensor`. The same shape as `tensor`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.rsqrt.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.rsqrt.md index 5f76fcd593..ba3787031e 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.rsqrt.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.rsqrt.md @@ -7,10 +7,10 @@ I.e., \\(y = 1 / \sqrt{x}\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_add.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_add.md index a8f8b7a9b0..883124f32c 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_add.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_add.md @@ -28,11 +28,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to add to `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the addition will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_div.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_div.md index ecd8e8b890..42e3089426 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_div.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.scatter_div.md @@ -24,11 +24,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of values that `ref` is divided by. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the operation will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.sparse_segment_sqrt_n.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.sparse_segment_sqrt_n.md index 83ae3d67ec..da9ca1f401 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.sparse_segment_sqrt_n.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.sparse_segment_sqrt_n.md @@ -11,16 +11,16 @@ of segments. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.subtract.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.subtract.md index 8108fb2054..f1bb91ef06 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.subtract.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.subtract.md @@ -8,11 +8,11 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.truncatemod.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.truncatemod.md index c75108fc55..9e57142bec 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.truncatemod.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.truncatemod.md @@ -11,11 +11,11 @@ with a flooring divide. E.g. `floor(x / y) * y + mod(x, y) = x`. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.zeta.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.zeta.md index ed66237d38..4ac4e1fc76 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.zeta.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.zeta.md @@ -11,11 +11,11 @@ The Hurwitz zeta function is defined as: ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`q`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`q`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.assign_add.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.assign_add.md index c57e4857d5..4c6b4a8ae7 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.assign_add.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.assign_add.md @@ -8,9 +8,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be added to the variable. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the addition will be protected by a lock; diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.fill.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.fill.md index 76de3e2d4d..1912fd8174 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.fill.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.fill.md @@ -15,9 +15,9 @@ fill([2, 3], 9) ==> [[9, 9, 9] ##### Args: -* <b>`dims`</b>: A `Tensor` of type `int32`. +* <b>`dims`</b>: An `Output` of type `int32`. 1-D. Represents the shape of the output tensor. -* <b>`value`</b>: A `Tensor`. 0-D (scalar). Value to fill the returned tensor. +* <b>`value`</b>: A `Output`. 0-D (scalar). Value to fill the returned tensor. @compatibility(numpy) Equivalent to np.full @@ -27,5 +27,5 @@ fill([2, 3], 9) ==> [[9, 9, 9] ##### Returns: - A `Tensor`. Has the same type as `value`. + A `Output`. Has the same type as `value`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.gather.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.gather.md index 3c6be5988c..ddb857f392 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.gather.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.gather.md @@ -26,12 +26,12 @@ this operation will permute `params` accordingly. ##### Args: -* <b>`params`</b>: A `Tensor`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`params`</b>: A `Output`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. * <b>`validate_indices`</b>: An optional `bool`. Defaults to `True`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `params`. + A `Output`. Has the same type as `params`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.image.encode_png.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.image.encode_png.md index fa073a771f..329825e75a 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.image.encode_png.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.image.encode_png.md @@ -17,12 +17,12 @@ the smallest output, but is slower. ##### Args: -* <b>`image`</b>: A `Tensor`. Must be one of the following types: `uint8`, `uint16`. +* <b>`image`</b>: A `Output`. Must be one of the following types: `uint8`, `uint16`. 3-D with shape `[height, width, channels]`. * <b>`compression`</b>: An optional `int`. Defaults to `-1`. Compression level. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. 0-D. PNG-encoded image. + An `Output` of type `string`. 0-D. PNG-encoded image. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_determinant.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_determinant.md index bcd0859e47..5960f9c130 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_determinant.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_determinant.md @@ -9,11 +9,11 @@ for all input submatrices `[..., :, :]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. Shape is `[..., M, M]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[...]`. + A `Output`. Has the same type as `input`. Shape is `[...]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_triangular_solve.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_triangular_solve.md index 66403eccfe..49c5543db0 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_triangular_solve.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_triangular_solve.md @@ -21,9 +21,9 @@ If `adjoint` is `False` then the strictly then the innermost matrices in ##### Args: -* <b>`matrix`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`matrix`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. -* <b>`rhs`</b>: A `Tensor`. Must have the same type as `matrix`. +* <b>`rhs`</b>: A `Output`. Must have the same type as `matrix`. Shape is `[..., M, K]`. * <b>`lower`</b>: An optional `bool`. Defaults to `True`. Boolean indicating whether the innermost matrices in `matrix` are @@ -40,5 +40,5 @@ If `adjoint` is `False` then the strictly then the innermost matrices in ##### Returns: - A `Tensor`. Has the same type as `matrix`. Shape is `[..., M, K]`. + A `Output`. Has the same type as `matrix`. Shape is `[..., M, K]`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.nn.max_pool_with_argmax.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.nn.max_pool_with_argmax.md index 5424efd7a7..9627ef4e0f 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.nn.max_pool_with_argmax.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.nn.max_pool_with_argmax.md @@ -9,7 +9,7 @@ The indices in `argmax` are flattened, so that a maximum value at position ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D with shape `[batch, height, width, channels]`. Input to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 4`. The size of the window for each dimension of the input tensor. @@ -23,8 +23,8 @@ The indices in `argmax` are flattened, so that a maximum value at position ##### Returns: - A tuple of `Tensor` objects (output, argmax). + A tuple of `Output` objects (output, argmax). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. The max pooled output tensor. -* <b>`argmax`</b>: A `Tensor` of type `Targmax`. 4-D. The flattened indices of the max values chosen for each output. +* <b>`output`</b>: A `Output`. Has the same type as `input`. The max pooled output tensor. +* <b>`argmax`</b>: A `Output` of type `Targmax`. 4-D. The flattened indices of the max values chosen for each output. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.realdiv.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.realdiv.md index facd6630dd..6f999ecd13 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.realdiv.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.realdiv.md @@ -10,11 +10,11 @@ If `x` and `y` are reals, this will return the floating-point division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.space_to_batch.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.space_to_batch.md index d83baacdd3..24a717b6e5 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.space_to_batch.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.space_to_batch.md @@ -13,8 +13,8 @@ block size. ##### Args: -* <b>`input`</b>: A `Tensor`. 4-D with shape `[batch, height, width, depth]`. -* <b>`paddings`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. 4-D with shape `[batch, height, width, depth]`. +* <b>`paddings`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies the padding of the input with zeros across the spatial dimensions as follows: @@ -106,5 +106,5 @@ block size. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.squared_difference.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.squared_difference.md index 19f25f473d..24d54be98d 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.squared_difference.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.squared_difference.md @@ -8,11 +8,11 @@ Returns (x - y)(x - y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.string_to_number.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.string_to_number.md index c6837bfa4a..8720aeafe8 100644 --- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.string_to_number.md +++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.string_to_number.md @@ -8,13 +8,13 @@ results in a rounded value.) ##### Args: -* <b>`string_tensor`</b>: A `Tensor` of type `string`. +* <b>`string_tensor`</b>: An `Output` of type `string`. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.float32, tf.int32`. Defaults to `tf.float32`. The numeric type to interpret each string in `string_tensor` as. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `out_type`. + A `Output` of type `out_type`. A Tensor of the same shape as the input `string_tensor`. diff --git a/tensorflow/g3doc/api_docs/python/image.md b/tensorflow/g3doc/api_docs/python/image.md index c08c4fa84f..11276d64d6 100644 --- a/tensorflow/g3doc/api_docs/python/image.md +++ b/tensorflow/g3doc/api_docs/python/image.md @@ -35,12 +35,12 @@ convert $src.gif -coalesce $dst.gif ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The GIF-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The GIF-encoded image. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `uint8`. + An `Output` of type `uint8`. 4-D with shape `[num_frames, height, width, 3]`. RGB order @@ -70,7 +70,7 @@ downscaling the image later. ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The JPEG-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The JPEG-encoded image. * <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for the decoded image. * <b>`ratio`</b>: An optional `int`. Defaults to `1`. Downscaling ratio. @@ -86,7 +86,7 @@ downscaling the image later. ##### Returns: - A `Tensor` of type `uint8`. 3-D with shape `[height, width, channels]`.. + An `Output` of type `uint8`. 3-D with shape `[height, width, channels]`.. - - - @@ -115,7 +115,7 @@ in function of the number of channels in `image`: ##### Args: -* <b>`image`</b>: A `Tensor` of type `uint8`. +* <b>`image`</b>: An `Output` of type `uint8`. 3-D with shape `[height, width, channels]`. * <b>`format`</b>: An optional `string` from: `"", "grayscale", "rgb"`. Defaults to `""`. Per pixel image format. @@ -140,7 +140,7 @@ in function of the number of channels in `image`: ##### Returns: - A `Tensor` of type `string`. 0-D. JPEG-encoded image. + An `Output` of type `string`. 0-D. JPEG-encoded image. @@ -166,7 +166,7 @@ of color channels. ##### Args: -* <b>`contents`</b>: A `Tensor` of type `string`. 0-D. The PNG-encoded image. +* <b>`contents`</b>: An `Output` of type `string`. 0-D. The PNG-encoded image. * <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for the decoded image. * <b>`dtype`</b>: An optional `tf.DType` from: `tf.uint8, tf.uint16`. Defaults to `tf.uint8`. @@ -174,7 +174,7 @@ of color channels. ##### Returns: - A `Tensor` of type `dtype`. 3-D with shape `[height, width, channels]`. + A `Output` of type `dtype`. 3-D with shape `[height, width, channels]`. - - - @@ -198,14 +198,14 @@ the smallest output, but is slower. ##### Args: -* <b>`image`</b>: A `Tensor`. Must be one of the following types: `uint8`, `uint16`. +* <b>`image`</b>: A `Output`. Must be one of the following types: `uint8`, `uint16`. 3-D with shape `[height, width, channels]`. * <b>`compression`</b>: An optional `int`. Defaults to `-1`. Compression level. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. 0-D. PNG-encoded image. + An `Output` of type `string`. 0-D. PNG-encoded image. @@ -287,7 +287,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -299,7 +299,7 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. @@ -314,7 +314,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -326,7 +326,7 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. @@ -341,7 +341,7 @@ Input images can be of different types but output images are always float. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -353,7 +353,7 @@ Input images can be of different types but output images are always float. ##### Returns: - A `Tensor` of type `float32`. 4-D with shape + An `Output` of type `float32`. 4-D with shape `[batch, new_height, new_width, channels]`. @@ -366,7 +366,7 @@ Resize `images` to `size` using nearest neighbor interpolation. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. 4-D with shape `[batch, height, width, channels]`. * <b>`size`</b>: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new size for the images. @@ -378,7 +378,7 @@ Resize `images` to `size` using nearest neighbor interpolation. ##### Returns: - A `Tensor`. Has the same type as `images`. 4-D with shape + A `Output`. Has the same type as `images`. 4-D with shape `[batch, new_height, new_width, channels]`. @@ -551,13 +551,13 @@ The argument `normalized` and `centered` controls how the windows are built: ##### Args: -* <b>`input`</b>: A `Tensor` of type `float32`. +* <b>`input`</b>: An `Output` of type `float32`. A 4-D float tensor of shape `[batch_size, height, width, channels]`. -* <b>`size`</b>: A `Tensor` of type `int32`. +* <b>`size`</b>: An `Output` of type `int32`. A 1-D tensor of 2 elements containing the size of the glimpses to extract. The glimpse height must be specified first, following by the glimpse width. -* <b>`offsets`</b>: A `Tensor` of type `float32`. +* <b>`offsets`</b>: An `Output` of type `float32`. A 2-D integer tensor of shape `[batch_size, 2]` containing the x, y locations of the center of each window. * <b>`centered`</b>: An optional `bool`. Defaults to `True`. @@ -574,7 +574,7 @@ The argument `normalized` and `centered` controls how the windows are built: ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. A tensor representing the glimpses `[batch_size, glimpse_height, glimpse_width, channels]`. @@ -598,10 +598,10 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. ##### Args: -* <b>`image`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. +* <b>`image`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`, `half`, `float32`, `float64`. A 4-D tensor of shape `[batch, image_height, image_width, depth]`. Both `image_height` and `image_width` need to be positive. -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor specifies the coordinates of a box in the `box_ind[i]` image and is specified in normalized coordinates `[y1, x1, y2, x2]`. A normalized coordinate value of @@ -612,10 +612,10 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. image. The width dimension is treated similarly. Normalized coordinates outside the `[0, 1]` range are allowed, in which case we use `extrapolation_value` to extrapolate the input image values. -* <b>`box_ind`</b>: A `Tensor` of type `int32`. +* <b>`box_ind`</b>: An `Output` of type `int32`. A 1-D tensor of shape `[num_boxes]` with int32 values in `[0, batch)`. The value of `box_ind[i]` specifies the image that the `i`-th box refers to. -* <b>`crop_size`</b>: A `Tensor` of type `int32`. +* <b>`crop_size`</b>: An `Output` of type `int32`. A 1-D tensor of 2 elements, `size = [crop_height, crop_width]`. All cropped image patches are resized to this size. The aspect ratio of the image content is not preserved. Both `crop_height` and `crop_width` need to be @@ -629,7 +629,7 @@ result is a 4-D tensor `[num_boxes, crop_height, crop_width, depth]`. ##### Returns: - A `Tensor` of type `float32`. + An `Output` of type `float32`. A 4-D tensor of shape `[num_boxes, crop_height, crop_width, depth]`. @@ -877,13 +877,13 @@ See `rgb_to_hsv` for a description of the HSV encoding. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. 1-D or higher rank. HSV data to convert. Last dimension must be size 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. `images` converted to RGB. + A `Output`. Has the same type as `images`. `images` converted to RGB. - - - @@ -903,13 +903,13 @@ corresponds to pure red, hue 1/3 is pure green, and 2/3 is pure blue. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. 1-D or higher rank. RGB data to convert. Last dimension must be size 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. `images` converted to HSV. + A `Output`. Has the same type as `images`. `images` converted to HSV. @@ -1283,16 +1283,16 @@ Parts of the bounding box may fall outside the image. ##### Args: -* <b>`images`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`images`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D with shape `[batch, height, width, depth]`. A batch of images. -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. 3-D with shape `[batch, num_bounding_boxes, 4]` containing bounding boxes. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `images`. + A `Output`. Has the same type as `images`. 4-D with the same shape as `images`. The batch of input images with bounding boxes drawn on the images. @@ -1325,12 +1325,12 @@ using the `tf.gather operation`. For example: ##### Args: -* <b>`boxes`</b>: A `Tensor` of type `float32`. +* <b>`boxes`</b>: An `Output` of type `float32`. A 2-D float tensor of shape `[num_boxes, 4]`. -* <b>`scores`</b>: A `Tensor` of type `float32`. +* <b>`scores`</b>: An `Output` of type `float32`. A 1-D float tensor of shape `[num_boxes]` representing a single score corresponding to each box (each row of boxes). -* <b>`max_output_size`</b>: A `Tensor` of type `int32`. +* <b>`max_output_size`</b>: An `Output` of type `int32`. A scalar integer tensor representing the maximum number of boxes to be selected by non max suppression. * <b>`iou_threshold`</b>: An optional `float`. Defaults to `0.5`. @@ -1340,7 +1340,7 @@ using the `tf.gather operation`. For example: ##### Returns: - A `Tensor` of type `int32`. + An `Output` of type `int32`. A 1-D integer tensor of shape `[M]` representing the selected indices from the boxes tensor, where `M <= max_output_size`. @@ -1393,9 +1393,9 @@ false and no bounding boxes are supplied, an error is raised. ##### Args: -* <b>`image_size`</b>: A `Tensor`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`. +* <b>`image_size`</b>: A `Output`. Must be one of the following types: `uint8`, `int8`, `int16`, `int32`, `int64`. 1-D, containing `[height, width, channels]`. -* <b>`bounding_boxes`</b>: A `Tensor` of type `float32`. +* <b>`bounding_boxes`</b>: An `Output` of type `float32`. 3-D with shape `[batch, N, 4]` describing the N bounding boxes associated with the image. * <b>`seed`</b>: An optional `int`. Defaults to `0`. @@ -1425,13 +1425,13 @@ false and no bounding boxes are supplied, an error is raised. ##### Returns: - A tuple of `Tensor` objects (begin, size, bboxes). + A tuple of `Output` objects (begin, size, bboxes). -* <b>`begin`</b>: A `Tensor`. Has the same type as `image_size`. 1-D, containing `[offset_height, offset_width, 0]`. Provide as input to +* <b>`begin`</b>: A `Output`. Has the same type as `image_size`. 1-D, containing `[offset_height, offset_width, 0]`. Provide as input to `tf.slice`. -* <b>`size`</b>: A `Tensor`. Has the same type as `image_size`. 1-D, containing `[target_height, target_width, -1]`. Provide as input to +* <b>`size`</b>: A `Output`. Has the same type as `image_size`. 1-D, containing `[target_height, target_width, -1]`. Provide as input to `tf.slice`. -* <b>`bboxes`</b>: A `Tensor` of type `float32`. 3-D with shape `[1, 1, 4]` containing the distorted bounding box. +* <b>`bboxes`</b>: An `Output` of type `float32`. 3-D with shape `[1, 1, 4]` containing the distorted bounding box. Provide as input to `tf.image.draw_bounding_boxes`. diff --git a/tensorflow/g3doc/api_docs/python/io_ops.md b/tensorflow/g3doc/api_docs/python/io_ops.md index 216586084a..1ba011ac89 100644 --- a/tensorflow/g3doc/api_docs/python/io_ops.md +++ b/tensorflow/g3doc/api_docs/python/io_ops.md @@ -59,14 +59,14 @@ A placeholder op that passes though `input` when its output is not fed. ##### Args: -* <b>`input`</b>: A `Tensor`. The default value to produce when `output` is not fed. +* <b>`input`</b>: A `Output`. The default value to produce when `output` is not fed. * <b>`shape`</b>: A `tf.TensorShape` or list of `ints`. The (possibly partial) shape of the tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. A placeholder tensor that defaults to `input` if it is not fed. @@ -1232,10 +1232,10 @@ Note that we allow leading and trailing spaces with int or float field. ##### Args: -* <b>`records`</b>: A `Tensor` of type `string`. +* <b>`records`</b>: An `Output` of type `string`. Each string is a record/row in the csv and all records should have the same format. -* <b>`record_defaults`</b>: A list of `Tensor` objects with types from: `float32`, `int32`, `int64`, `string`. +* <b>`record_defaults`</b>: A list of `Output` objects with types from: `float32`, `int32`, `int64`, `string`. One tensor per column of the input record, with either a scalar default value for that column or empty if the column is required. * <b>`field_delim`</b>: An optional `string`. Defaults to `","`. @@ -1244,7 +1244,7 @@ Note that we allow leading and trailing spaces with int or float field. ##### Returns: - A list of `Tensor` objects. Has the same type as `record_defaults`. + A list of `Output` objects. Has the same type as `record_defaults`. Each tensor will have the same shape as records. @@ -1257,7 +1257,7 @@ Reinterpret the bytes of a string as a vector of numbers. ##### Args: -* <b>`bytes`</b>: A `Tensor` of type `string`. +* <b>`bytes`</b>: An `Output` of type `string`. All the elements must have the same length. * <b>`out_type`</b>: A `tf.DType` from: `tf.half, tf.float32, tf.float64, tf.int32, tf.uint8, tf.int16, tf.int8, tf.int64`. * <b>`little_endian`</b>: An optional `bool`. Defaults to `True`. @@ -1268,7 +1268,7 @@ Reinterpret the bytes of a string as a vector of numbers. ##### Returns: - A `Tensor` of type `out_type`. + A `Output` of type `out_type`. A Tensor with one more dimension than the input `bytes`. The added dimension will have size equal to the length of the elements of `bytes` divided by the number of bytes to represent `out_type`. @@ -1660,7 +1660,7 @@ Transforms a serialized tensorflow.TensorProto proto into a Tensor. ##### Args: -* <b>`serialized`</b>: A `Tensor` of type `string`. +* <b>`serialized`</b>: An `Output` of type `string`. A scalar string containing a serialized TensorProto proto. * <b>`out_type`</b>: A `tf.DType`. The type of the serialized tensor. The provided type must match the @@ -1669,7 +1669,7 @@ Transforms a serialized tensorflow.TensorProto proto into a Tensor. ##### Returns: - A `Tensor` of type `out_type`. A Tensor of type `out_type`. + A `Output` of type `out_type`. A Tensor of type `out_type`. - - - @@ -1688,14 +1688,14 @@ Example-parsing ops. ##### Args: -* <b>`json_examples`</b>: A `Tensor` of type `string`. +* <b>`json_examples`</b>: An `Output` of type `string`. Each string is a JSON object serialized according to the JSON mapping of the Example proto. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. Each string is a binary Example protocol buffer corresponding to the respective element of `json_examples`. @@ -2699,12 +2699,12 @@ basename portion of the pattern, not in the directory portion. ##### Args: -* <b>`pattern`</b>: A `Tensor` of type `string`. A (scalar) shell wildcard pattern. +* <b>`pattern`</b>: An `Output` of type `string`. A (scalar) shell wildcard pattern. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. A vector of matching filenames. + An `Output` of type `string`. A vector of matching filenames. - - - @@ -2716,12 +2716,12 @@ Reads and outputs the entire contents of the input filename. ##### Args: -* <b>`filename`</b>: A `Tensor` of type `string`. +* <b>`filename`</b>: An `Output` of type `string`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. - - - @@ -2733,9 +2733,9 @@ Writes contents to the file at input filename. Creates file if not existing. ##### Args: -* <b>`filename`</b>: A `Tensor` of type `string`. +* <b>`filename`</b>: An `Output` of type `string`. scalar. The name of the file to which we write the contents. -* <b>`contents`</b>: A `Tensor` of type `string`. +* <b>`contents`</b>: An `Output` of type `string`. scalar. The content to be written to the output file. * <b>`name`</b>: A name for the operation (optional). diff --git a/tensorflow/g3doc/api_docs/python/math_ops.md b/tensorflow/g3doc/api_docs/python/math_ops.md index f70edc7423..9672599f8c 100644 --- a/tensorflow/g3doc/api_docs/python/math_ops.md +++ b/tensorflow/g3doc/api_docs/python/math_ops.md @@ -27,13 +27,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -48,13 +48,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -69,13 +69,13 @@ Returns x * y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -116,13 +116,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -215,13 +215,13 @@ If `x` and `y` are reals, this will return the floating-point division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -241,13 +241,13 @@ Python Semantics. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -262,13 +262,13 @@ Returns x // y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -286,13 +286,13 @@ with a flooring divide. E.g. `floor(x / y) * y + mod(x, y) = x`. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -310,13 +310,13 @@ with a flooring divide. E.g. `floor(x / y) * y + mod(x, y) = x`. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -331,13 +331,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -353,15 +353,15 @@ of corresponding 3-element vectors is cross-multiplied independently. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. A tensor containing 3-element vectors. -* <b>`b`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`b`</b>: A `Output`. Must have the same type as `a`. Another tensor, of same type and shape as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. Pairwise cross product of the vectors in `a` and `b`. @@ -405,7 +405,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -432,12 +433,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -473,12 +474,12 @@ I.e., \\(y = 1 / x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -558,12 +559,12 @@ I.e., \\(y = 1 / \sqrt{x}\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -604,12 +605,12 @@ Computes exponential of x element-wise. \\(y = e^x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -623,12 +624,12 @@ I.e., \\(y = \log_e x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -642,12 +643,12 @@ I.e., \\(y = \log_e (1 + x)\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -659,12 +660,12 @@ Returns element-wise smallest integer in not less than x. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -676,12 +677,12 @@ Returns element-wise largest integer not greater than x. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -696,13 +697,13 @@ Returns the max of x and y (i.e. x > y ? x : y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -717,13 +718,13 @@ Returns the min of x and y (i.e. x < y ? x : y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -735,12 +736,12 @@ Computes cos of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -752,12 +753,12 @@ Computes sin of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -803,12 +804,12 @@ Computes tan of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -820,12 +821,12 @@ Computes acos of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -837,12 +838,12 @@ Computes asin of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -854,12 +855,12 @@ Computes atan of x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -871,12 +872,12 @@ Computes the log of the absolute value of `Gamma(x)` element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -890,12 +891,12 @@ Computes Psi, the derivative of Lgamma (the log of the absolute value of ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -925,12 +926,12 @@ Computes the complementary error function of `x` element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -945,13 +946,13 @@ Returns (x - y)(x - y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -977,13 +978,13 @@ Gamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. - - - @@ -1009,13 +1010,13 @@ Gamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. - - - @@ -1033,13 +1034,13 @@ The Hurwitz zeta function is defined as: ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`q`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`q`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1058,13 +1059,13 @@ where \\(\psi(x)\\) is the digamma function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. - - - @@ -1090,14 +1091,14 @@ beta function. ##### Args: -* <b>`a`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`b`</b>: A `Tensor`. Must have the same type as `a`. -* <b>`x`</b>: A `Tensor`. Must have the same type as `a`. +* <b>`a`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`b`</b>: A `Output`. Must have the same type as `a`. +* <b>`x`</b>: A `Output`. Must have the same type as `a`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `a`. + A `Output`. Has the same type as `a`. @@ -1133,13 +1134,13 @@ tf.diag(diagonal) ==> [[1, 0, 0, 0] ##### Args: -* <b>`diagonal`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`diagonal`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. Rank k tensor where k is at most 3. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `diagonal`. + A `Output`. Has the same type as `diagonal`. - - - @@ -1170,13 +1171,13 @@ tf.diag_part(input) ==> [1, 2, 3, 4] ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. Rank k tensor where k is 2, 4, or 6. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. The extracted diagonal. + A `Output`. Has the same type as `input`. The extracted diagonal. - - - @@ -1351,12 +1352,12 @@ which has shape (2, 4, 4) ##### Args: -* <b>`diagonal`</b>: A `Tensor`. Rank `k`, where `k >= 1`. +* <b>`diagonal`</b>: A `Output`. Rank `k`, where `k >= 1`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `diagonal`. + A `Output`. Has the same type as `diagonal`. Rank `k+1`, with `output.shape = diagonal.shape + [diagonal.shape[-1]]`. @@ -1398,12 +1399,12 @@ which has shape (2, 4) ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k` tensor where `k >= 2`. +* <b>`input`</b>: A `Output`. Rank `k` tensor where `k >= 2`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The extracted diagonal(s) having shape `diagonal.shape = input.shape[:-2] + [min(input.shape[-2:])]`. @@ -1457,18 +1458,18 @@ Useful special cases: ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k` tensor. -* <b>`num_lower`</b>: A `Tensor` of type `int64`. +* <b>`input`</b>: A `Output`. Rank `k` tensor. +* <b>`num_lower`</b>: An `Output` of type `int64`. 0-D tensor. Number of subdiagonals to keep. If negative, keep entire lower triangle. -* <b>`num_upper`</b>: A `Tensor` of type `int64`. +* <b>`num_upper`</b>: An `Output` of type `int64`. 0-D tensor. Number of superdiagonals to keep. If negative, keep entire upper triangle. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Rank `k` tensor of the same shape as input. The extracted banded tensor. @@ -1494,14 +1495,14 @@ tensor of rank `k+1` with dimensions `[I, J, K, ..., M, N]` where: ##### Args: -* <b>`input`</b>: A `Tensor`. Rank `k+1`, where `k >= 1`. -* <b>`diagonal`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Rank `k+1`, where `k >= 1`. +* <b>`diagonal`</b>: A `Output`. Must have the same type as `input`. Rank `k`, where `k >= 1`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. Rank `k+1`, with `output.shape = input.shape`. @@ -1545,22 +1546,25 @@ tf.matrix_transpose(x) ==> [[1 4] - - - -### `tf.matmul(a, b, transpose_a=False, transpose_b=False, a_is_sparse=False, b_is_sparse=False, name=None)` {#matmul} +### `tf.matmul(a, b, transpose_a=False, transpose_b=False, adjoint_a=False, adjoint_b=False, a_is_sparse=False, b_is_sparse=False, name=None)` {#matmul} Multiplies matrix `a` by matrix `b`, producing `a` * `b`. -The inputs must be two-dimensional matrices, with matching inner dimensions, -possibly after transposition. +The inputs must be matrices (or tensors of rank > 2, representing batches of +matrices), with matching inner dimensions, possibly after transposition. Both matrices must be of the same type. The supported types are: -`float32`, `float64`, `int32`, `complex64`. +`float16`, `float32`, `float64`, `int32`, `complex64`, `complex128`. -Either matrix can be transposed on the fly by setting the corresponding flag -to `True`. This is `False` by default. +Either matrix can be transposed or adjointed (conjugated and transposed) on +the fly by setting one of the corresponding flag to `True`. These are `False` +by default. If one or both of the matrices contain a lot of zeros, a more efficient multiplication algorithm can be used by setting the corresponding `a_is_sparse` or `b_is_sparse` flag to `True`. These are `False` by default. +This optimization is only available for plain matrices (rank-2 tensors) with +datatypes `bfloat16` or `float32`. For example: @@ -1574,22 +1578,57 @@ b = tf.constant([7, 8, 9, 10, 11, 12], shape=[3, 2]) => [[7. 8.] [11. 12.]] c = tf.matmul(a, b) => [[58 64] [139 154]] + + +# 3-D tensor `a` +a = tf.constant(np.arange(1,13), shape=[2, 2, 3]) => [[[ 1. 2. 3.] + [ 4. 5. 6.]], + [[ 7. 8. 9.] + [10. 11. 12.]]] + +# 3-D tensor `b` +b = tf.constant(np.arange(13,25), shape=[2, 3, 2]) => [[[13. 14.] + [15. 16.] + [17. 18.]], + [[19. 20.] + [21. 22.] + [23. 24.]]] +c = tf.matmul(a, b) => [[[ 94 100] + [229 244]], + [[508 532] + [697 730]]] ``` ##### Args: -* <b>`a`</b>: `Tensor` of type `float32`, `float64`, `int32` or `complex64`. -* <b>`b`</b>: `Tensor` with same type as `a`. +* <b>`a`</b>: `Tensor` of type `float16`, `float32`, `float64`, `int32`, `complex64`, + `complex128` and rank > 1. +* <b>`b`</b>: `Tensor` with same type and rank as `a`. * <b>`transpose_a`</b>: If `True`, `a` is transposed before multiplication. * <b>`transpose_b`</b>: If `True`, `b` is transposed before multiplication. +* <b>`adjoint_a`</b>: If `True`, `a` is conjugated and transposed before + multiplication. +* <b>`adjoint_b`</b>: If `True`, `b` is conjugated and transposed before + multiplication. * <b>`a_is_sparse`</b>: If `True`, `a` is treated as a sparse matrix. * <b>`b_is_sparse`</b>: If `True`, `b` is treated as a sparse matrix. * <b>`name`</b>: Name for the operation (optional). ##### Returns: - A `Tensor` of the same type as `a`. + A `Tensor` of the same type as `a` and `b` where each inner-most matrix is + the product of the corresponding matrices in `a` and `b, e.g. if all + transpose or adjoint attributes are `False`: + + output[..., :, :] = a[..., :, :] * b[..., :, :] , + + +##### Raises: + + +* <b>`ValueError`</b>: If transpose_a and adjoint_a, or transpose_b and adjoint_b + are both set to True. - - - @@ -1598,7 +1637,7 @@ c = tf.matmul(a, b) => [[58 64] Multiplies slices of two tensors in batches. -Multiplies all slices of `Tensor` `x` and `y` (each slice can be +Multiplies all slices of `Output` `x` and `y` (each slice can be viewed as an element of a batch), and arranges the individual results in a single output tensor of the same batch size. Each of the individual slices can optionally be adjointed (to adjoint a matrix @@ -1620,9 +1659,9 @@ It is computed as: ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `complex64`, `complex128`. 3-D or higher with shape `[..., r_x, c_x]`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. 3-D or higher with shape `[..., r_y, c_y]`. * <b>`adj_x`</b>: An optional `bool`. Defaults to `False`. If `True`, adjoint the slices of `x`. Defaults to `False`. @@ -1632,7 +1671,7 @@ It is computed as: ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. 3-D or higher with shape `[..., r_o, c_o]` @@ -1650,13 +1689,13 @@ for all input submatrices `[..., :, :]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. Shape is `[..., M, M]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[...]`. + A `Output`. Has the same type as `input`. Shape is `[...]`. - - - @@ -1680,14 +1719,14 @@ garbage result. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. * <b>`adjoint`</b>: An optional `bool`. Defaults to `False`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[..., M, M]`. + A `Output`. Has the same type as `input`. Shape is `[..., M, M]`. @compatibility(numpy) Equivalent to np.linalg.inv @@ -1708,13 +1747,13 @@ containing the Cholesky decompositions for all input submatrices `[..., :, :]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. Shape is `[..., M, M]`. + A `Output`. Has the same type as `input`. Shape is `[..., M, M]`. - - - @@ -1771,9 +1810,9 @@ If `adjoint` is `True` then each output matrix satisfies ##### Args: -* <b>`matrix`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`, `complex64`, `complex128`. +* <b>`matrix`</b>: A `Output`. Must be one of the following types: `float64`, `float32`, `complex64`, `complex128`. Shape is `[..., M, M]`. -* <b>`rhs`</b>: A `Tensor`. Must have the same type as `matrix`. +* <b>`rhs`</b>: A `Output`. Must have the same type as `matrix`. Shape is `[..., M, K]`. * <b>`adjoint`</b>: An optional `bool`. Defaults to `False`. Boolean indicating whether to solve with `matrix` or its (block-wise) @@ -1782,7 +1821,7 @@ If `adjoint` is `True` then each output matrix satisfies ##### Returns: - A `Tensor`. Has the same type as `matrix`. Shape is `[..., M, K]`. + A `Output`. Has the same type as `matrix`. Shape is `[..., M, K]`. - - - @@ -1810,9 +1849,9 @@ If `adjoint` is `False` then the strictly then the innermost matrices in ##### Args: -* <b>`matrix`</b>: A `Tensor`. Must be one of the following types: `float64`, `float32`. +* <b>`matrix`</b>: A `Output`. Must be one of the following types: `float64`, `float32`. Shape is `[..., M, M]`. -* <b>`rhs`</b>: A `Tensor`. Must have the same type as `matrix`. +* <b>`rhs`</b>: A `Output`. Must have the same type as `matrix`. Shape is `[..., M, K]`. * <b>`lower`</b>: An optional `bool`. Defaults to `True`. Boolean indicating whether the innermost matrices in `matrix` are @@ -1829,7 +1868,7 @@ If `adjoint` is `False` then the strictly then the innermost matrices in ##### Returns: - A `Tensor`. Has the same type as `matrix`. Shape is `[..., M, K]`. + A `Output`. Has the same type as `matrix`. Shape is `[..., M, K]`. - - - @@ -2105,7 +2144,8 @@ tf.imag(input) ==> [4.75, 5.75] ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `complex64`, `complex128`. +* <b>`input`</b>: A `Tensor`. Must be one of the following types: `complex64`, + `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: @@ -2162,12 +2202,12 @@ dimension of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most dimension of `input` is replaced with its 1D Fourier Transform. @@ -2183,12 +2223,12 @@ dimension of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most dimension of `input` is replaced with its inverse 1D Fourier Transform. @@ -2204,12 +2244,12 @@ Compute the 2-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 2 dimensions of `input` are replaced with their 2D Fourier Transform. @@ -2229,12 +2269,12 @@ Compute the inverse 2-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 2 dimensions of `input` are replaced with their inverse 2D Fourier Transform. @@ -2254,12 +2294,12 @@ dimensions of `input`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 3 dimensions of `input` are replaced with their 3D Fourier Transform. @@ -2279,12 +2319,12 @@ Compute the inverse 3-dimensional discrete Fourier Transform over the inner-most ##### Args: -* <b>`input`</b>: A `Tensor` of type `complex64`. A complex64 tensor. +* <b>`input`</b>: An `Output` of type `complex64`. A complex64 tensor. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `complex64`. + An `Output` of type `complex64`. A complex64 tensor of the same shape as `input`. The inner-most 3 dimensions of `input` are replaced with their inverse 3D Fourier Transform. @@ -2923,15 +2963,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -2957,15 +2997,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -2991,15 +3031,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3024,15 +3064,15 @@ that `segment_ids[j] == i`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3059,15 +3099,15 @@ values summed. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor whose rank is equal to the rank of `data`'s first dimension. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3100,15 +3140,15 @@ If the sum is empty for a given segment ID `i`, `output[i] = 0`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`segment_ids`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`segment_ids`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor whose shape is a prefix of `data.shape`. -* <b>`num_segments`</b>: A `Tensor` of type `int32`. +* <b>`num_segments`</b>: An `Output` of type `int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for the first `segment_ids.rank` dimensions, which are replaced with a single dimension which has size `num_segments`. @@ -3154,16 +3194,16 @@ tf.segment_sum(c, tf.constant([0, 0, 1])) ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3184,16 +3224,16 @@ dimension, selecting a subset of dimension 0, specified by `indices`. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3213,16 +3253,16 @@ of segments. ##### Args: -* <b>`data`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`data`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A 1-D tensor. Has same rank as `segment_ids`. -* <b>`segment_ids`</b>: A `Tensor` of type `int32`. +* <b>`segment_ids`</b>: An `Output` of type `int32`. A 1-D tensor. Values should be sorted and can be repeated. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `data`. + A `Output`. Has the same type as `data`. Has same shape as data, except for dimension 0 which has size `k`, the number of segments. @@ -3245,15 +3285,15 @@ Returns the index with the smallest value across axiss of a tensor. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. int32, 0 <= axis < rank(input). Describes which axis of the input Tensor to reduce across. For vectors, use axis = 0. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. - - - @@ -3265,15 +3305,15 @@ Returns the index with the largest value across axiss of a tensor. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. -* <b>`axis`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`axis`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. int32, 0 <= axis < rank(input). Describes which axis of the input Tensor to reduce across. For vectors, use axis = 0. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. @@ -3308,17 +3348,17 @@ idx ==> [1, 3, 5] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. Values to keep. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. 1-D. Values to remove. +* <b>`x`</b>: A `Output`. 1-D. Values to keep. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. 1-D. Values to remove. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (out, idx). + A tuple of `Output` objects (out, idx). -* <b>`out`</b>: A `Tensor`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. +* <b>`out`</b>: A `Output`. Has the same type as `x`. 1-D. Values present in `x` but not in `y`. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. Positions of `x` values preserved in `out`. - - - @@ -3396,16 +3436,16 @@ idx ==> [0, 0, 1, 2, 2, 2, 3, 4, 4] ##### Args: -* <b>`x`</b>: A `Tensor`. 1-D. +* <b>`x`</b>: A `Output`. 1-D. * <b>`out_idx`</b>: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (y, idx). + A tuple of `Output` objects (y, idx). -* <b>`y`</b>: A `Tensor`. Has the same type as `x`. 1-D. -* <b>`idx`</b>: A `Tensor` of type `out_idx`. 1-D. +* <b>`y`</b>: A `Output`. Has the same type as `x`. 1-D. +* <b>`idx`</b>: A `Output` of type `out_idx`. 1-D. @@ -3503,12 +3543,12 @@ invert_permutation(x) ==> [2, 4, 3, 0, 1] ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. 1-D. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. 1-D. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. 1-D. + A `Output`. Has the same type as `x`. 1-D. @@ -3525,13 +3565,13 @@ Returns x * y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -3566,12 +3606,12 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. diff --git a/tensorflow/g3doc/api_docs/python/nn.md b/tensorflow/g3doc/api_docs/python/nn.md index 7344190ab7..14f50e0892 100644 --- a/tensorflow/g3doc/api_docs/python/nn.md +++ b/tensorflow/g3doc/api_docs/python/nn.md @@ -27,12 +27,12 @@ Computes rectified linear: `max(features, 0)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. - - - @@ -88,12 +88,12 @@ See [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs) ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. - - - @@ -105,12 +105,12 @@ Computes softplus: `log(exp(features) + 1)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. - - - @@ -122,12 +122,12 @@ Computes softsign: `features / (abs(features) + 1)`. ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `features`. + A `Output`. Has the same type as `features`. - - - @@ -461,8 +461,8 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. * <b>`strides`</b>: A list of `ints`. 1-D of length 4. The stride of the sliding window for each dimension of `input`. Must be in the same order as the dimension specified with format. @@ -479,7 +479,7 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -801,9 +801,9 @@ Our Conv3D implements a form of cross-correlation. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, in_depth, in_height, in_width, in_channels]`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. Shape `[filter_depth, filter_height, filter_width, in_channels, out_channels]`. `in_channels` must match between `input` and `filter`. * <b>`strides`</b>: A list of `ints` that has length `>= 5`. @@ -815,7 +815,7 @@ Our Conv3D implements a form of cross-correlation. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -939,7 +939,7 @@ The indices in `argmax` are flattened, so that a maximum value at position ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D with shape `[batch, height, width, channels]`. Input to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 4`. The size of the window for each dimension of the input tensor. @@ -953,10 +953,10 @@ The indices in `argmax` are flattened, so that a maximum value at position ##### Returns: - A tuple of `Tensor` objects (output, argmax). + A tuple of `Output` objects (output, argmax). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. The max pooled output tensor. -* <b>`argmax`</b>: A `Tensor` of type `Targmax`. 4-D. The flattened indices of the max values chosen for each output. +* <b>`output`</b>: A `Output`. Has the same type as `input`. The max pooled output tensor. +* <b>`argmax`</b>: A `Output` of type `Targmax`. 4-D. The flattened indices of the max values chosen for each output. - - - @@ -968,7 +968,7 @@ Performs 3D average pooling on the input. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, depth, rows, cols, channels]` tensor to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 5`. 1-D tensor of length 5. The size of the window for each dimension of @@ -982,7 +982,7 @@ Performs 3D average pooling on the input. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. The average pooled output tensor. @@ -995,7 +995,7 @@ Performs 3D max pooling on the input. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Shape `[batch, depth, rows, cols, channels]` tensor to pool over. * <b>`ksize`</b>: A list of `ints` that has length `>= 5`. 1-D tensor of length 5. The size of the window for each dimension of @@ -1009,7 +1009,7 @@ Performs 3D max pooling on the input. ##### Returns: - A `Tensor`. Has the same type as `input`. The max pooled output tensor. + A `Output`. Has the same type as `input`. The max pooled output tensor. - - - @@ -1026,7 +1026,7 @@ pooling region. ##### Args: -* <b>`value`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. +* <b>`value`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. 4-D with shape `[batch, height, width, channels]`. * <b>`pooling_ratio`</b>: A list of `floats` that has length `>= 4`. Pooling ratio for each dimension of `value`, currently only @@ -1065,11 +1065,11 @@ pooling region. ##### Returns: - A tuple of `Tensor` objects (output, row_pooling_sequence, col_pooling_sequence). + A tuple of `Output` objects (output, row_pooling_sequence, col_pooling_sequence). -* <b>`output`</b>: A `Tensor`. Has the same type as `value`. output tensor after fractional avg pooling. -* <b>`row_pooling_sequence`</b>: A `Tensor` of type `int64`. row pooling sequence, needed to calculate gradient. -* <b>`col_pooling_sequence`</b>: A `Tensor` of type `int64`. column pooling sequence, needed to calculate gradient. +* <b>`output`</b>: A `Output`. Has the same type as `value`. output tensor after fractional avg pooling. +* <b>`row_pooling_sequence`</b>: An `Output` of type `int64`. row pooling sequence, needed to calculate gradient. +* <b>`col_pooling_sequence`</b>: An `Output` of type `int64`. column pooling sequence, needed to calculate gradient. - - - @@ -1110,7 +1110,7 @@ For more details on fractional max pooling, see this paper: ##### Args: -* <b>`value`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. +* <b>`value`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`. 4-D with shape `[batch, height, width, channels]`. * <b>`pooling_ratio`</b>: A list of `floats` that has length `>= 4`. Pooling ratio for each dimension of `value`, currently only @@ -1149,11 +1149,11 @@ For more details on fractional max pooling, see this paper: ##### Returns: - A tuple of `Tensor` objects (output, row_pooling_sequence, col_pooling_sequence). + A tuple of `Output` objects (output, row_pooling_sequence, col_pooling_sequence). -* <b>`output`</b>: A `Tensor`. Has the same type as `value`. output tensor after fractional max pooling. -* <b>`row_pooling_sequence`</b>: A `Tensor` of type `int64`. row pooling sequence, needed to calculate gradient. -* <b>`col_pooling_sequence`</b>: A `Tensor` of type `int64`. column pooling sequence, needed to calculate gradient. +* <b>`output`</b>: A `Output`. Has the same type as `value`. output tensor after fractional max pooling. +* <b>`row_pooling_sequence`</b>: An `Output` of type `int64`. row pooling sequence, needed to calculate gradient. +* <b>`col_pooling_sequence`</b>: An `Output` of type `int64`. column pooling sequence, needed to calculate gradient. - - - @@ -1308,9 +1308,9 @@ negation of the erosion of `-input` by the reflected `filter`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 4-D with shape `[batch, in_height, in_width, depth]`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. 3-D with shape `[filter_height, filter_width, depth]`. * <b>`strides`</b>: A list of `ints` that has length `>= 4`. The stride of the sliding window for each dimension of the input @@ -1324,7 +1324,7 @@ negation of the erosion of `-input` by the reflected `filter`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. 4-D with shape `[batch, out_height, out_width, depth]`. @@ -1438,7 +1438,7 @@ convolutional neural networks (NIPS 2012)](http://papers.nips.cc/paper/4824-imag ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `half`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `half`. 4-D. * <b>`depth_radius`</b>: An optional `int`. Defaults to `5`. 0-D. Half-width of the 1-D normalization window. @@ -1451,7 +1451,7 @@ convolutional neural networks (NIPS 2012)](http://papers.nips.cc/paper/4824-imag ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. - - - @@ -1584,13 +1584,13 @@ Computes half the L2 norm of a tensor without the `sqrt`: ##### Args: -* <b>`t`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`t`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Typically 2-D, but may have any dimensions. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `t`. 0-D. + A `Output`. Has the same type as `t`. 0-D. - - - @@ -2827,16 +2827,16 @@ $$out_i = predictions_{i, targets_i} \in TopKIncludingTies(predictions_i)$$ ##### Args: -* <b>`predictions`</b>: A `Tensor` of type `float32`. +* <b>`predictions`</b>: An `Output` of type `float32`. A `batch_size` x `classes` tensor. -* <b>`targets`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`targets`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A `batch_size` vector of class ids. * <b>`k`</b>: An `int`. Number of top elements to look at for computing precision. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. Computed Precision at `k` as a `bool Tensor`. + An `Output` of type `bool`. Computed Precision at `k` as a `bool Tensor`. @@ -3282,22 +3282,22 @@ Computes Quantized Rectified Linear X: `min(max(features, 0), max_value)` ##### Args: -* <b>`features`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. -* <b>`max_value`</b>: A `Tensor` of type `float32`. -* <b>`min_features`</b>: A `Tensor` of type `float32`. +* <b>`features`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`max_value`</b>: An `Output` of type `float32`. +* <b>`min_features`</b>: An `Output` of type `float32`. The float value that the lowest quantized value represents. -* <b>`max_features`</b>: A `Tensor` of type `float32`. +* <b>`max_features`</b>: An `Output` of type `float32`. The float value that the highest quantized value represents. * <b>`out_type`</b>: An optional `tf.DType` from: `tf.qint8, tf.quint8, tf.qint16, tf.quint16, tf.qint32`. Defaults to `tf.quint8`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A tuple of `Tensor` objects (activations, min_activations, max_activations). + A tuple of `Output` objects (activations, min_activations, max_activations). -* <b>`activations`</b>: A `Tensor` of type `out_type`. Has the same output shape as "features". -* <b>`min_activations`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized value represents. -* <b>`max_activations`</b>: A `Tensor` of type `float32`. The float value that the highest quantized value represents. +* <b>`activations`</b>: A `Output` of type `out_type`. Has the same output shape as "features". +* <b>`min_activations`</b>: An `Output` of type `float32`. The float value that the lowest quantized value represents. +* <b>`max_activations`</b>: An `Output` of type `float32`. The float value that the highest quantized value represents. - - - @@ -3309,11 +3309,11 @@ Produces the max pool of the input tensor for quantized types. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. The 4D (batch x rows x cols x depth) Tensor to MaxReduce over. -* <b>`min_input`</b>: A `Tensor` of type `float32`. +* <b>`min_input`</b>: An `Output` of type `float32`. The float value that the lowest quantized input value represents. -* <b>`max_input`</b>: A `Tensor` of type `float32`. +* <b>`max_input`</b>: An `Output` of type `float32`. The float value that the highest quantized input value represents. * <b>`ksize`</b>: A list of `ints`. The size of the window for each dimension of the input tensor. @@ -3327,11 +3327,11 @@ Produces the max pool of the input tensor for quantized types. ##### Returns: - A tuple of `Tensor` objects (output, min_output, max_output). + A tuple of `Output` objects (output, min_output, max_output). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. -* <b>`min_output`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized output value represents. -* <b>`max_output`</b>: A `Tensor` of type `float32`. The float value that the highest quantized output value represents. +* <b>`output`</b>: A `Output`. Has the same type as `input`. +* <b>`min_output`</b>: An `Output` of type `float32`. The float value that the lowest quantized output value represents. +* <b>`max_output`</b>: An `Output` of type `float32`. The float value that the highest quantized output value represents. - - - @@ -3343,11 +3343,11 @@ Produces the average pool of the input tensor for quantized types. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `qint8`, `quint8`, `qint16`, `quint16`, `qint32`. 4-D with shape `[batch, height, width, channels]`. -* <b>`min_input`</b>: A `Tensor` of type `float32`. +* <b>`min_input`</b>: An `Output` of type `float32`. The float value that the lowest quantized input value represents. -* <b>`max_input`</b>: A `Tensor` of type `float32`. +* <b>`max_input`</b>: An `Output` of type `float32`. The float value that the highest quantized input value represents. * <b>`ksize`</b>: A list of `ints`. The size of the window for each dimension of the input tensor. @@ -3361,11 +3361,11 @@ Produces the average pool of the input tensor for quantized types. ##### Returns: - A tuple of `Tensor` objects (output, min_output, max_output). + A tuple of `Output` objects (output, min_output, max_output). -* <b>`output`</b>: A `Tensor`. Has the same type as `input`. -* <b>`min_output`</b>: A `Tensor` of type `float32`. The float value that the lowest quantized output value represents. -* <b>`max_output`</b>: A `Tensor` of type `float32`. The float value that the highest quantized output value represents. +* <b>`output`</b>: A `Output`. Has the same type as `input`. +* <b>`min_output`</b>: An `Output` of type `float32`. The float value that the lowest quantized output value represents. +* <b>`max_output`</b>: An `Output` of type `float32`. The float value that the highest quantized output value represents. @@ -3446,8 +3446,8 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`. -* <b>`filter`</b>: A `Tensor`. Must have the same type as `input`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `float32`, `float64`. +* <b>`filter`</b>: A `Output`. Must have the same type as `input`. * <b>`strides`</b>: A list of `ints`. 1-D of length 4. The stride of the sliding window for each dimension of `input`. @@ -3457,6 +3457,6 @@ horizontal and vertices strides, `strides = [1, stride, stride, 1]`. ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. diff --git a/tensorflow/g3doc/api_docs/python/sparse_ops.md b/tensorflow/g3doc/api_docs/python/sparse_ops.md index 2590b21fcc..48e3b70a59 100644 --- a/tensorflow/g3doc/api_docs/python/sparse_ops.md +++ b/tensorflow/g3doc/api_docs/python/sparse_ops.md @@ -169,20 +169,20 @@ the other direction. ##### Args: -* <b>`sp_indices`</b>: A `Tensor` of type `int64`. +* <b>`sp_indices`</b>: An `Output` of type `int64`. 2-D. `N x R` matrix with the indices of non-empty values in a SparseTensor, possibly not in canonical ordering. -* <b>`sp_values`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`sp_values`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 1-D. `N` non-empty values corresponding to `sp_indices`. -* <b>`sp_shape`</b>: A `Tensor` of type `int64`. +* <b>`sp_shape`</b>: An `Output` of type `int64`. 1-D. Shape of the input SparseTensor. -* <b>`dense`</b>: A `Tensor`. Must have the same type as `sp_values`. +* <b>`dense`</b>: A `Output`. Must have the same type as `sp_values`. `R`-D. The dense Tensor operand. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `sp_values`. + A `Output`. Has the same type as `sp_values`. 1-D. The `N` values that are operated on. @@ -202,20 +202,20 @@ the other direction. ##### Args: -* <b>`sp_indices`</b>: A `Tensor` of type `int64`. +* <b>`sp_indices`</b>: An `Output` of type `int64`. 2-D. `N x R` matrix with the indices of non-empty values in a SparseTensor, possibly not in canonical ordering. -* <b>`sp_values`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`sp_values`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 1-D. `N` non-empty values corresponding to `sp_indices`. -* <b>`sp_shape`</b>: A `Tensor` of type `int64`. +* <b>`sp_shape`</b>: An `Output` of type `int64`. 1-D. Shape of the input SparseTensor. -* <b>`dense`</b>: A `Tensor`. Must have the same type as `sp_values`. +* <b>`dense`</b>: A `Output`. Must have the same type as `sp_values`. `R`-D. The dense Tensor operand. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `sp_values`. + A `Output`. Has the same type as `sp_values`. 1-D. The `N` values that are operated on. diff --git a/tensorflow/g3doc/api_docs/python/state_ops.md b/tensorflow/g3doc/api_docs/python/state_ops.md index d1d5045ee7..9debad8771 100644 --- a/tensorflow/g3doc/api_docs/python/state_ops.md +++ b/tensorflow/g3doc/api_docs/python/state_ops.md @@ -407,7 +407,8 @@ containing the absolute value of each element in `x`. For example, if x is an input element and y is an output element, this operation computes \\(y = |x|\\). -See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex +See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a +complex number. ##### Args: @@ -435,13 +436,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -456,13 +457,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -477,13 +478,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -533,13 +534,13 @@ Returns the truth value of (x >= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -605,13 +606,13 @@ Returns the truth value of (x > y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -623,12 +624,12 @@ Returns the truth value of NOT x element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -659,13 +660,13 @@ Returns the truth value of (x <= y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -680,13 +681,13 @@ Returns the truth value of (x < y) element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -701,13 +702,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -728,12 +729,12 @@ I.e., \\(y = -x\\). ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -748,13 +749,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -798,13 +799,13 @@ Returns x + y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -819,13 +820,13 @@ Returns the truth value of x AND y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -840,13 +841,13 @@ Returns x / y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `uint16`, `int16`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -896,13 +897,13 @@ Returns element-wise remainder of division. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -924,13 +925,13 @@ Returns the truth value of x OR y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor` of type `bool`. -* <b>`y`</b>: A `Tensor` of type `bool`. +* <b>`x`</b>: An `Output` of type `bool`. +* <b>`y`</b>: An `Output` of type `bool`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `bool`. + An `Output` of type `bool`. - - - @@ -974,13 +975,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1043,13 +1044,13 @@ Returns x - y element-wise. ##### Args: -* <b>`x`</b>: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. -* <b>`y`</b>: A `Tensor`. Must have the same type as `x`. +* <b>`x`</b>: A `Output`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. +* <b>`y`</b>: A `Output`. Must have the same type as `x`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `x`. + A `Output`. Has the same type as `x`. - - - @@ -1408,9 +1409,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. +* <b>`ref`</b>: A mutable `Output`. Should be from a `Variable` node. May be uninitialized. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be assigned to the variable. * <b>`validate_shape`</b>: An optional `bool`. Defaults to `True`. If true, the operation will validate that the shape @@ -1439,9 +1440,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be added to the variable. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the addition will be protected by a lock; @@ -1466,9 +1467,9 @@ This makes it easier to chain operations that need to use the reset value. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`value`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`value`</b>: A `Output`. Must have the same type as `ref`. The value to be subtracted to the variable. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the subtraction will be protected by a lock; @@ -2764,10 +2765,10 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. Should be from a `Variable` node. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to store in `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `True`. If True, the assignment will be protected by a lock; @@ -2812,11 +2813,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to add to `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the addition will be protected by a lock; @@ -2859,11 +2860,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to subtract from `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the subtraction will be protected by a lock; @@ -2904,11 +2905,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of updated values to multiply to `ref`. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the operation will be protected by a lock; @@ -2949,11 +2950,11 @@ Requires `updates.shape = indices.shape + ref.shape[1:]`. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be from a `Variable` node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A tensor of values that `ref` is divided by. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. If True, the operation will be protected by a lock; @@ -2974,7 +2975,7 @@ Applies sparse `updates` to individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -2983,7 +2984,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -3009,11 +3010,11 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`ref`</b>: A mutable `Output`. A mutable Tensor. Should be from a Variable node. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to add to ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `True`. @@ -3024,7 +3025,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. @@ -3037,7 +3038,7 @@ Applies sparse addition between `updates` and individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -3046,7 +3047,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -3072,12 +3073,12 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to add to ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. @@ -3088,7 +3089,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. @@ -3101,7 +3102,7 @@ Applies sparse subtraction between `updates` and individual values or slices within a given variable according to `indices`. -`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`. +`ref` is an `Output` with rank `P` and `indices` is an `Output` of rank `Q`. `indices` must be integer tensor, containing indices into `ref`. It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`. @@ -3110,7 +3111,7 @@ The innermost dimension of `indices` (with length `K`) corresponds to indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th dimension of `ref`. -`updates` is `Tensor` of rank `Q-1+P-K` with shape: +`updates` is `Output` of rank `Q-1+P-K` with shape: ``` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. @@ -3136,12 +3137,12 @@ slices. ##### Args: -* <b>`ref`</b>: A mutable `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. +* <b>`ref`</b>: A mutable `Output`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. A mutable Tensor. Should be from a Variable node. -* <b>`indices`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`indices`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. A Tensor. Must be one of the following types: int32, int64. A tensor of indices into ref. -* <b>`updates`</b>: A `Tensor`. Must have the same type as `ref`. +* <b>`updates`</b>: A `Output`. Must have the same type as `ref`. A Tensor. Must have the same type as ref. A tensor of updated values to subtract from ref. * <b>`use_locking`</b>: An optional `bool`. Defaults to `False`. @@ -3152,7 +3153,7 @@ slices. ##### Returns: - A mutable `Tensor`. Has the same type as `ref`. + A mutable `Output`. Has the same type as `ref`. Same as ref. Returned as a convenience for operations that want to use the updated values after the update is done. diff --git a/tensorflow/g3doc/api_docs/python/string_ops.md b/tensorflow/g3doc/api_docs/python/string_ops.md index 86878ca664..460d3dd20b 100644 --- a/tensorflow/g3doc/api_docs/python/string_ops.md +++ b/tensorflow/g3doc/api_docs/python/string_ops.md @@ -28,13 +28,13 @@ to the same bucket. To prevent this problem, use a strong hash function with ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. The strings to assign a hash bucket. +* <b>`input`</b>: An `Output` of type `string`. The strings to assign a hash bucket. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. @@ -58,7 +58,7 @@ time than `tf.string_to_hash_bucket_fast`. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. The strings to assign a hash bucket. +* <b>`input`</b>: An `Output` of type `string`. The strings to assign a hash bucket. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`key`</b>: A list of `ints`. The key for the keyed hash function passed as a list of two uint64 @@ -67,7 +67,7 @@ time than `tf.string_to_hash_bucket_fast`. ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. @@ -87,13 +87,13 @@ This functionality will be deprecated and it's recommended to use ##### Args: -* <b>`string_tensor`</b>: A `Tensor` of type `string`. +* <b>`string_tensor`</b>: An `Output` of type `string`. * <b>`num_buckets`</b>: An `int` that is `>= 1`. The number of buckets. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `int64`. + An `Output` of type `int64`. A Tensor of the same shape as the input `string_tensor`. @@ -136,9 +136,9 @@ tf.reduce_join(a, []) ==> ["abcd"] ##### Args: -* <b>`inputs`</b>: A `Tensor` of type `string`. +* <b>`inputs`</b>: An `Output` of type `string`. The input to be joined. All reduced indices must have non-zero size. -* <b>`reduction_indices`</b>: A `Tensor` of type `int32`. +* <b>`reduction_indices`</b>: An `Output` of type `int32`. The dimensions to reduce over. Dimensions are reduced in the order specified. Omitting `reduction_indices` is equivalent to passing `[n-1, n-2, ..., 0]`. Negative indices from `-n` to `-1` are supported. @@ -150,7 +150,7 @@ tf.reduce_join(a, []) ==> ["abcd"] ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. Has shape equal to that of the input with reduced dimensions removed or set to `1` depending on `keep_dims`. @@ -166,7 +166,7 @@ with the given separator (default is an empty separator). ##### Args: -* <b>`inputs`</b>: A list of at least 1 `Tensor` objects of type `string`. +* <b>`inputs`</b>: A list of at least 1 `Output` objects of type `string`. A list of string tensors. The tensors must all have the same shape, or be scalars. Scalars may be mixed in; these will be broadcast to the shape of non-scalar inputs. @@ -176,7 +176,7 @@ with the given separator (default is an empty separator). ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. @@ -231,9 +231,9 @@ st.values = ['hello', 'world', 'a', 'b', 'c'] ### `tf.substr(input, pos, len, name=None)` {#substr} -Return substrings from `Tensor` of strings. +Return substrings from `Output` of strings. -For each string in the input `Tensor`, creates a substring starting at index +For each string in the input `Output`, creates a substring starting at index `pos` with a total length of `len`. If `len` defines a substring that would extend beyond the length of the input @@ -310,16 +310,16 @@ output = [b'hir', b'ee', b'n"] ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Tensor of strings -* <b>`pos`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`. +* <b>`input`</b>: An `Output` of type `string`. Tensor of strings +* <b>`pos`</b>: A `Output`. Must be one of the following types: `int32`, `int64`. Scalar defining the position of first character in each substring -* <b>`len`</b>: A `Tensor`. Must have the same type as `pos`. +* <b>`len`</b>: A `Output`. Must have the same type as `pos`. Scalar defining the number of characters to include in each substring * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Tensor of substrings + An `Output` of type `string`. Tensor of substrings @@ -336,7 +336,7 @@ types and boolean. ##### Args: -* <b>`input`</b>: A `Tensor`. Must be one of the following types: `int32`, `int64`, `complex64`, `float32`, `float64`, `bool`, `int8`. +* <b>`input`</b>: A `Output`. Must be one of the following types: `int32`, `int64`, `complex64`, `float32`, `float64`, `bool`, `int8`. * <b>`precision`</b>: An optional `int`. Defaults to `-1`. The post-decimal precision to use for floating point numbers. Only used if precision > -1. @@ -356,7 +356,7 @@ types and boolean. ##### Returns: - A `Tensor` of type `string`. + An `Output` of type `string`. - - - @@ -375,14 +375,14 @@ Web-safe means that the encoder uses - and _ instead of + and /. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Strings to be encoded. +* <b>`input`</b>: An `Output` of type `string`. Strings to be encoded. * <b>`pad`</b>: An optional `bool`. Defaults to `False`. Bool whether padding is applied at the ends. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Input strings encoded in base64. + An `Output` of type `string`. Input strings encoded in base64. - - - @@ -397,11 +397,11 @@ Web-safe means that input must use - and _ instead of + and /. ##### Args: -* <b>`input`</b>: A `Tensor` of type `string`. Base64 strings to decode. +* <b>`input`</b>: An `Output` of type `string`. Base64 strings to decode. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor` of type `string`. Decoded strings. + An `Output` of type `string`. Decoded strings. diff --git a/tensorflow/g3doc/api_docs/python/train.md b/tensorflow/g3doc/api_docs/python/train.md index 0d45989723..4190739489 100644 --- a/tensorflow/g3doc/api_docs/python/train.md +++ b/tensorflow/g3doc/api_docs/python/train.md @@ -735,12 +735,12 @@ to pretend that the value was a constant. Some examples include: ##### Args: -* <b>`input`</b>: A `Tensor`. +* <b>`input`</b>: A `Output`. * <b>`name`</b>: A name for the operation (optional). ##### Returns: - A `Tensor`. Has the same type as `input`. + A `Output`. Has the same type as `input`. |