| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Change `reduce_logsumexp` to internally use `reshape` rather than `squeeze`
since the latter requires the `axis` arg to be a Python `list`.
PiperOrigin-RevId: 183396533
* Kernel utils to support broadcast add and mul.
PiperOrigin-RevId: 183397494
* Updating sparsify_gather.
PiperOrigin-RevId: 183402917
* [tf.data] Move slow-path-related code into the slow path in IteratorHandleOp::Compute().
This slightly reduces the amount of work performed when an iterator is accessed (after the first access), and potentially reduces contention if concurrent steps are accessing the same iterator.
PiperOrigin-RevId: 183406221
* Cleanup: Ran clang-format on all *.{cc,h} in under grappler.
PiperOrigin-RevId: 183406440
* Increase shard count of //third_party/tensorflow/python:nn_batchnorm_test to avoid timeouts
When run under asan, the test runs for about 5 minutes, and sometimes
longer, causing frequent timeouts.
This change increases the shard count of the test to 4, which brings the run time
of the longest running shard under asan to about 2 minutes.
PiperOrigin-RevId: 183414888
* Add available choices to toco flags and fix minor formatting issues.
PiperOrigin-RevId: 183415713
* Performance improvements to some GPU code to use shared locks instead of unique locks for some hotspot cases.
PiperOrigin-RevId: 183418559
* [XLA] Improve error message for bad slices.
PiperOrigin-RevId: 183420038
* Fix py3 build rules for all py tests under py2tf.
PiperOrigin-RevId: 183422144
* Fix bug with Operation._control_inputs setter.
PiperOrigin-RevId: 183422192
* Make softmax_op_test.py work with C API enabled.
PiperOrigin-RevId: 183422829
* Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/core/kernels.
PiperOrigin-RevId: 183423961
* Fix the documentation for the dense layer for how rank > 2 inputs are handled.
PiperOrigin-RevId: 183425868
* Cleanup: Ran clang-format on all *.{cc,h} in tensorflow/core/ops.
PiperOrigin-RevId: 183429339
|
|
|
These new operations are more general versions of the existing
SpaceToBatch and BatchToSpace operations that support an arbitrary
number of spatial dimensions and a non-uniform block shape.
For backwards compatibility, the existing SpaceToBatch and BatchToSpace
operations are retained, but internally forward to the new, generic
implementation.
For convenience, the new operations support an arbitrary number of
trailing non-spatial dimensions, rather than requiring a single batch
dimension.
As an optimization, leading and trailing spatial dimensions with no
padding and a block size of 1 are collapsed such that they incur no
additional overhead.
The maximum number of (non-collapsed) spatial dimensions must be
specified at compile time in spacetobatch_functor.h, and is currently
set to 4.
Change: 133836914
|
Rasmus Munk Larsen
A. Unique TensorFlower