| Commit message (Collapse) | Author | Age |
|
|
|
| |
PiperOrigin-RevId: 214321627
|
|
|
|
|
|
| |
operators that need it.
PiperOrigin-RevId: 214320700
|
|
|
|
| |
PiperOrigin-RevId: 214311663
|
|
|
|
| |
PiperOrigin-RevId: 214309598
|
|
|
|
| |
PiperOrigin-RevId: 214309210
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also a small bugfix to handle unknown shapes in backprop._num_elements.
Before:
entry {
name: "L2hmcBenchmark.eager_train_cpu_defun"
iters: 10
wall_time: 0.594115018845
extras {
key: "examples_per_sec"
value {
double_value: 336.635152548
}
}
}
After:
entry {
name: "L2hmcBenchmark.eager_train_cpu_defun"
iters: 10
wall_time: 0.322251081467
extras {
key: "examples_per_sec"
value {
double_value: 620.634069216
}
}
}
PiperOrigin-RevId: 214308142
|
|
|
|
|
|
|
|
|
|
| |
CudnnConvParams, just pass around the HloInstruction.
This is based on the observation that most code doesn't care about the
convolution semantics like which operand is input vs filter vs output.
In fact, only layout assignment and conv runner care about them.
PiperOrigin-RevId: 214307399
|
|
|
|
|
|
|
|
| |
self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about:
* the fact that the session may be reused.
* the session is not closed even when doing a "with self.test_session()" statement.
PiperOrigin-RevId: 214300210
|
|
|
|
|
|
|
| |
This allows the Keras learning phase to work inside functions and defuns.
Note: There might still be bugs in graph mode if the default placeholder is being fed (instead of using set_learning_phase) and a layer is in a function.
PiperOrigin-RevId: 214299002
|
|
|
|
| |
PiperOrigin-RevId: 214298224
|
|\
| |
| |
| | |
PiperOrigin-RevId: 214297796
|
| |
| |
| |
| | |
PiperOrigin-RevId: 214296771
|
| |
| |
| |
| | |
PiperOrigin-RevId: 214295534
|
| |
| |
| |
| |
| |
| | |
tf.contrib.data.
PiperOrigin-RevId: 214294955
|
| |
| |
| |
| | |
PiperOrigin-RevId: 214294594
|
|\ \
| | |
| | |
| | | |
PiperOrigin-RevId: 214290400
|
|\ \ \
| | | |
| | | |
| | | | |
PiperOrigin-RevId: 214289882
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
that uses circular references, which can be problematic in eager mode.
PiperOrigin-RevId: 214287432
|
| | | |
| | | |
| | | |
| | | | |
PiperOrigin-RevId: 214286888
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
self.session(graph=...) as it's the same semantic.
PiperOrigin-RevId: 214286845
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
These properties are necessary for the strategy to work with
`tf.estimator.train_and_evaluate`.
PiperOrigin-RevId: 214285957
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Output names for custom ops and properly insert newlines for
null IntVectors.
PiperOrigin-RevId: 214285696
|
| | | |
| | | |
| | | |
| | | | |
PiperOrigin-RevId: 214279868
|
|\ \ \ \
| | | | |
| | | | |
| | | | | |
PiperOrigin-RevId: 214278672
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
forward compatibility window has expired.
PiperOrigin-RevId: 214277870
|
| | | | |
| | | | |
| | | | |
| | | | | |
PiperOrigin-RevId: 214275960
|
|\ \ \ \ \
| | | | | |
| | | | | |
| | | | | | |
PiperOrigin-RevId: 214271324
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214269484
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
about the expected functionality.
PiperOrigin-RevId: 214266947
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214263489
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
example).
PiperOrigin-RevId: 214262549
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
and vice versa.
PiperOrigin-RevId: 214257872
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214255941
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214251264
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
It only worked for ranks 1 or 2, and only if the dimension to sort is the most minor dimension.
Also fix the SafeLess function so that the SortExtremeValues() test passes.
PiperOrigin-RevId: 214239560
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214232622
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
PiperOrigin-RevId: 214230777
|
|\ \ \ \ \ \ \
| | | | | | | |
| | | | | | | |
| | | | | | | | |
PiperOrigin-RevId: 214221868
|
|\ \ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
PiperOrigin-RevId: 214220618
|
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
PiperOrigin-RevId: 214219282
|
|\ \ \ \ \ \ \ \ \
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
PiperOrigin-RevId: 214215589
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Fix for b/116169724
Only remove noops if they refer to const nodes.
PiperOrigin-RevId: 214199200
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
This change switches `tf.contrib.data.Optional` to use a `Structure` class to represent
the structure of its value, instead of `output_types`, `output_shapes`, and `output_classes` properties. It adds support for nesting `Optional` objects and representing their structure.
This change also makes a modification to the `Structure` class: `Structure.is_compatible_with(x)` now takes another `Structure` as the `x` argument, instead of a value. This makes it easier to work with nested structures (where we might not have a value readily available), and better matches the interface of other `is_compatible_with()` methods (e.g. in `tf.TensorShape` and `tf.DType`).
Finally, in the process of making this change, I observed possible crash-failures when a DT_VARIANT tensor containing another DT_VARIANT tensor is copied between CPU and GPU. This change "fixes" the immediate problem by raising an UnimplementedError, but more work will be necessary to support the full range of use cases.
PiperOrigin-RevId: 214198993
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
PiperOrigin-RevId: 214195020
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
PiperOrigin-RevId: 214179792
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
MD fetching method.
TFRecord files do not contain a file-level header that describes the
MD of the file. To avoid backwards compatibility issues, we add a
lightweight function that computes the statistics over the file once
and caches the result for future calls.
A future implementor could do a better job of computing GetMetadata() by
having the RecordWriter emit these entries during writing so that
GetMetadata() only reads. Doing so will require additional backwards
compatibility checks to ensure that the function works both for old
TFRecords and the new format.
PiperOrigin-RevId: 214178704
|
|\ \ \ \ \ \ \ \ \ \
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | | |
PiperOrigin-RevId: 214177065
|
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | | |
PiperOrigin-RevId: 214176089
|
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | | |
PiperOrigin-RevId: 214174935
|
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | |
| | | | | | | | | | | |
PiperOrigin-RevId: 214173896
|