Eugene Brevdo [Thu, 21 Dec 2017 07:02:11 +0000 (23:02 -0800)]
[tf nest] Add additional key yielder.
This is just copying a utility function created by Malcolm Reynolds.
PiperOrigin-RevId:
179775504
Derek Murray [Thu, 21 Dec 2017 07:01:10 +0000 (23:01 -0800)]
Revert change to ParseBytesList that failed on some platforms.
PiperOrigin-RevId:
179775354
A. Unique TensorFlower [Thu, 21 Dec 2017 05:15:44 +0000 (21:15 -0800)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
179769852
Anna R [Thu, 21 Dec 2017 05:06:48 +0000 (21:06 -0800)]
Internal change.
PiperOrigin-RevId:
179769382
A. Unique TensorFlower [Thu, 21 Dec 2017 05:04:31 +0000 (21:04 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
179769207
Derek Murray [Thu, 21 Dec 2017 04:54:25 +0000 (20:54 -0800)]
[tf.data] Add `tf.contrib.data.parse_single_example()`.
The new op is a fused implementation of the existing
`tf.parse_single_example()`, which is more efficient when parsing a
single Example at a time.
PiperOrigin-RevId:
179768512
Anna R [Thu, 21 Dec 2017 04:17:15 +0000 (20:17 -0800)]
Automated g4 rollback of changelist
179757188
PiperOrigin-RevId:
179766237
A. Unique TensorFlower [Thu, 21 Dec 2017 03:01:09 +0000 (19:01 -0800)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
179761889
Anna R [Thu, 21 Dec 2017 02:59:24 +0000 (18:59 -0800)]
Automated g4 rollback of changelist
179735577
PiperOrigin-RevId:
179761787
A. Unique TensorFlower [Thu, 21 Dec 2017 01:59:07 +0000 (17:59 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
179757188
Chris Leary [Thu, 21 Dec 2017 01:55:13 +0000 (17:55 -0800)]
[XLA] Add CrossReplicaSum to Python API bindings.
Add support to CPU for the trivial replica_count=1 case.
PiperOrigin-RevId:
179756956
Justin Lebar [Thu, 21 Dec 2017 01:54:39 +0000 (17:54 -0800)]
[XLA] Fix HLO graph dumper not to assume that instruction names start with "%".
HLO graph dumper needs to be aware that we've gotten rid of the "%"
prefix in HLO names so it doesn't print e.g.
reduce
reduce.42
Subcomputation: add
...
but instead simply prints
reduce.42
Subcomputation: add
...
PiperOrigin-RevId:
179756922
A. Unique TensorFlower [Thu, 21 Dec 2017 01:53:22 +0000 (17:53 -0800)]
Fix typo in struct name: RecvNodeDescritorHash -> RecvNodeDescriptorHash.
PiperOrigin-RevId:
179756840
Yangzihao Wang [Thu, 21 Dec 2017 01:48:42 +0000 (17:48 -0800)]
Fix the condition check for int8 fused_conv so that it works on Volta.
PiperOrigin-RevId:
179756490
Roy Frostig [Thu, 21 Dec 2017 01:08:43 +0000 (17:08 -0800)]
[XLA] Add Collapse and Rev ops to the local XLA Python client.
PiperOrigin-RevId:
179753056
Alexandre Passos [Thu, 21 Dec 2017 00:43:33 +0000 (16:43 -0800)]
implicit_gradients always sorts its return values by variable creation order
PiperOrigin-RevId:
179750392
Yao Zhang [Thu, 21 Dec 2017 00:13:59 +0000 (16:13 -0800)]
Support max pool v2.
PiperOrigin-RevId:
179747281
Anna R [Wed, 20 Dec 2017 23:34:58 +0000 (15:34 -0800)]
Automated g4 rollback of changelist
179729641
PiperOrigin-RevId:
179742670
Brennan Saeta [Wed, 20 Dec 2017 22:47:04 +0000 (14:47 -0800)]
Truncate status messages before converting to gRPC.
PiperOrigin-RevId:
179736746
Alexandre Passos [Wed, 20 Dec 2017 22:37:26 +0000 (14:37 -0800)]
Implements the fill op for int64 indices.
PiperOrigin-RevId:
179735577
Peter Hawkins [Wed, 20 Dec 2017 22:32:34 +0000 (14:32 -0800)]
[TF:XLA] Move XlaResource into its own file, and refactor it into a better-abstracted class. No functional changes intended.
PiperOrigin-RevId:
179734920
Justin Lebar [Wed, 20 Dec 2017 22:13:37 +0000 (14:13 -0800)]
[XLA] Add reassociation rule for adds, and add canonicalization rules for add/sub of a constant.
This patch adds a new algebraic simplification:
* (A + C0) + C1 => A + (C0 + C1), where C0 and C1 are constants. This
allows us to constant-fold C0 + C1.
In service of this rule, this patch also adds two new canonicalizations:
* Const + A => A + Const
* A - Const => A + (-1 * Const)
PiperOrigin-RevId:
179731747
A. Unique TensorFlower [Wed, 20 Dec 2017 22:01:11 +0000 (14:01 -0800)]
Adds a randomized test for Atan.
Resolves TODOs that are out of scope because conv op registration is non-C64 only.
PiperOrigin-RevId:
179730049
A. Unique TensorFlower [Wed, 20 Dec 2017 21:57:22 +0000 (13:57 -0800)]
Add the capability of verbose logging for HTTP requests to the TensorFlow GCS client.
This change allows the GCS client within TensorFlow to log verbose information about HTTP requests to the standard Google logger. (Actually, to the TensorFlow variant of the standard Google logger.)
This capability is disabled by default. It is enabled by setting the environment variable GCS_LOG_HTTP_REQUEST_VERBOSE=1 (or =true).
PiperOrigin-RevId:
179729641
Austin Anderson [Wed, 20 Dec 2017 21:53:00 +0000 (13:53 -0800)]
Make tf_cc_test compilable by default for Android
This change adjusts the build options for tf_cc_test targets so that
they can be built for Android devices by default. Many targets will
still need additional options on a per-target basis.
Android requires that executables be compiled with `-pie`.
PiperOrigin-RevId:
179729092
A. Unique TensorFlower [Wed, 20 Dec 2017 21:31:06 +0000 (13:31 -0800)]
Destroy batch_thread_pool_ first. For the new in flight batches implementation, the callbacks scheduled on this thread pool refer to other class members which must remain alive until the thread pool is empty.
Also fix a similar lifetime issue in the unit tests.
PiperOrigin-RevId:
179726389
Yao Zhang [Wed, 20 Dec 2017 21:30:22 +0000 (13:30 -0800)]
Extend DataFormatDimMap to handle tensors.
PiperOrigin-RevId:
179726269
Brennan Saeta [Wed, 20 Dec 2017 21:29:07 +0000 (13:29 -0800)]
Add prefetching into parallel_interleave
This change adds 2 parameters to parallel_interleave:
- prefetch_input_elements: determines the number of iterators to prefetch
allowing buffers to warm up and data to be pre-fetched without blocking
the main thread (i.e. the GetNext() call).
- buffer_output_elements: in order to avoid creating thousands of threads, we
fuse in the .prefetch() operator as an additional parameter. The value of
this parameter is identical to the value passed to `.prefetch()`
PiperOrigin-RevId:
179726088
Peter Hawkins [Wed, 20 Dec 2017 21:26:43 +0000 (13:26 -0800)]
Restore escaping to string attribute list summaries accidentally removed by previous change.
PiperOrigin-RevId:
179725715
Yao Zhang [Wed, 20 Dec 2017 21:26:37 +0000 (13:26 -0800)]
Support SplitV op.
PiperOrigin-RevId:
179725704
Chris Leary [Wed, 20 Dec 2017 21:15:49 +0000 (13:15 -0800)]
[XLA] Remove inappropriate plugin linkage.
PiperOrigin-RevId:
179724373
Russell Power [Wed, 20 Dec 2017 20:58:46 +0000 (12:58 -0800)]
Fix logging of excessive unsupported operations.
An operator precedence issue with the previous version triggered errors when models had a large number of summaries.
PiperOrigin-RevId:
179722697
A. Unique TensorFlower [Wed, 20 Dec 2017 19:10:40 +0000 (11:10 -0800)]
Convert Status-based call sequence checks to CHECK-based.
Call sequence requirements are correctness requirements; any violation of a correctness requirement is a code defect / bug. Currently, some code in the TF GCS client uses Status objects to report call sequence violations. This tolerates bugs, allowing them to go undetected and uncorrected.
This change converts many call sequence checks to use CHECK, rather than Status. The result is code that is simpler, and that exposes bugs, rather than protecting them.
PiperOrigin-RevId:
179709047
A. Unique TensorFlower [Wed, 20 Dec 2017 18:57:08 +0000 (10:57 -0800)]
Adding mention of contrib optimizers and new learning rate decays.
PiperOrigin-RevId:
179706954
A. Unique TensorFlower [Wed, 20 Dec 2017 18:31:55 +0000 (10:31 -0800)]
Adding an optimizer module for ConstantSGD.
PiperOrigin-RevId:
179703560
A. Unique TensorFlower [Wed, 20 Dec 2017 18:17:28 +0000 (10:17 -0800)]
Adding missing comma in __init__.py and fixing links in NOS optimizers docs.
PiperOrigin-RevId:
179701465
Russell Power [Wed, 20 Dec 2017 18:10:32 +0000 (10:10 -0800)]
Reduce the number of warning messages for unsupported operations.
PiperOrigin-RevId:
179700417
Yao Zhang [Wed, 20 Dec 2017 18:02:12 +0000 (10:02 -0800)]
Support more control flow ops. Delete ops that require ref inputs (e.g.
variables); these ops are not applicable to layout optimizer, because
layout-agnostic nodes are never directly connected to variables, but to
other layout-sensitive nodes such as conv and pooling.
PiperOrigin-RevId:
179699124
Allen Lavoie [Wed, 20 Dec 2017 17:50:56 +0000 (09:50 -0800)]
Add some unit tests for memory optimizer _recompute_hint Python usage (mostly
for documentation purposes).
PiperOrigin-RevId:
179697882
Jingyue Wu [Wed, 20 Dec 2017 16:34:54 +0000 (08:34 -0800)]
Fix a flaky test.
In the old code, `a` can be freed before bytes_in_use, making the result of
`bytes_in_use` unstable. In the new code, `b` uses `a` and control-depends on
`bytes_in_use`, so `a` cannot be freed before `bytes_in_use`.
PiperOrigin-RevId:
179690026
A. Unique TensorFlower [Wed, 20 Dec 2017 15:38:32 +0000 (07:38 -0800)]
Add support for constant resolution of the Fill op.
PiperOrigin-RevId:
179684795
Mark Daoust [Wed, 20 Dec 2017 15:24:29 +0000 (07:24 -0800)]
rename files
PiperOrigin-RevId:
179683700
A. Unique TensorFlower [Wed, 20 Dec 2017 14:47:44 +0000 (06:47 -0800)]
Replace std::pow with MathUtil::IPow when the exponent is a non-negative integer.
Includes minor test cleanup.
PiperOrigin-RevId:
179680628
A. Unique TensorFlower [Wed, 20 Dec 2017 14:46:40 +0000 (06:46 -0800)]
Remove unused BUILD dependencies
PiperOrigin-RevId:
179680552
Raghuraman Krishnamoorthi [Wed, 20 Dec 2017 04:38:51 +0000 (20:38 -0800)]
PiperOrigin-RevId:
179639611
A. Unique TensorFlower [Wed, 20 Dec 2017 13:35:03 +0000 (05:35 -0800)]
[TF:XLA] Move ShardingBuilder out of computation_builder.h
ShardingBuilder does not belong in computation_builder.h - its presence there is puzzling and unintuitive.
PiperOrigin-RevId:
179675257
Neal Wu [Wed, 20 Dec 2017 03:56:58 +0000 (19:56 -0800)]
Specify that only NHWC is supported for certain ops on CPU specifically, rather than in general.
PiperOrigin-RevId:
179637312
Peter Hawkins [Wed, 20 Dec 2017 03:22:44 +0000 (19:22 -0800)]
Elide long strings and lists when summarizing AttrValues.
PiperOrigin-RevId:
179635492
Jianwei Xie [Wed, 20 Dec 2017 03:22:44 +0000 (19:22 -0800)]
Adds some more distributed training related fields into TF_CONFIG.
PiperOrigin-RevId:
179635491
Yao Zhang [Wed, 20 Dec 2017 03:09:00 +0000 (19:09 -0800)]
Support ternary ops.
PiperOrigin-RevId:
179634713
A. Unique TensorFlower [Wed, 20 Dec 2017 03:04:57 +0000 (19:04 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
179634365
A. Unique TensorFlower [Wed, 20 Dec 2017 02:36:51 +0000 (18:36 -0800)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
179632513
A. Unique TensorFlower [Wed, 20 Dec 2017 02:15:52 +0000 (18:15 -0800)]
Minor cleanup, now that there's a single Executable::ExecuteOnStream method.
PiperOrigin-RevId:
179630890
A. Unique TensorFlower [Wed, 20 Dec 2017 01:54:19 +0000 (17:54 -0800)]
[XLA:GPU] Make the use of scratch allocator in convolution_thunk safe.
Add member function Thunk::ShouldFutureScheduledThunksDependOn for
convolution_thunk to tell thunk executor that all future scheduled thunks
should wait for convolution_thunk. This can ensure that the use of scratch
allocator in convolution_thunk is safe.
PiperOrigin-RevId:
179628764
Anna R [Wed, 20 Dec 2017 01:28:06 +0000 (17:28 -0800)]
Read ApiDef from TensorFlow Go API.
PiperOrigin-RevId:
179625412
Derek Murray [Wed, 20 Dec 2017 01:26:02 +0000 (17:26 -0800)]
[tf.data] Add a transformation for getting the unique elements of a dataset.
PiperOrigin-RevId:
179625211
Roy Frostig [Wed, 20 Dec 2017 01:20:17 +0000 (17:20 -0800)]
Implement convolution operations in the local Python XLA client.
PiperOrigin-RevId:
179624737
Shanqing Cai [Wed, 20 Dec 2017 01:13:50 +0000 (17:13 -0800)]
tfdbg: send graph traceback and source files from TensorBoard wrapper and hook
To this end:
* Refactor some common, shared constants and methods to a new file: common.py
PiperOrigin-RevId:
179624165
Alexandre Passos [Wed, 20 Dec 2017 00:51:25 +0000 (16:51 -0800)]
TFE copy across gpus
PiperOrigin-RevId:
179622259
A. Unique TensorFlower [Wed, 20 Dec 2017 00:31:31 +0000 (16:31 -0800)]
K-FAC: Cross Replica Mean for TPU
Adds an op for taking the average of a Tensor across all TPU cores, and uses it
before updating covariance statistics. This is a no-op if TPUs aren't used.
PiperOrigin-RevId:
179620193
Justin Lebar [Wed, 20 Dec 2017 00:27:34 +0000 (16:27 -0800)]
[XLA] Rename CreateFullWithMonotonicDim0MajorLayout to CreateFullWithDescendingLayout.
We've renamed MonotonicDim0MajorLayout to DescendingLayout.
PiperOrigin-RevId:
179619563
A. Unique TensorFlower [Tue, 19 Dec 2017 23:50:58 +0000 (15:50 -0800)]
Add comparison ufuncs for numpy bfloat16 type.
Fix 2 bfloat16 tests.
PiperOrigin-RevId:
179614898
A. Unique TensorFlower [Tue, 19 Dec 2017 23:45:38 +0000 (15:45 -0800)]
[XLA] Fix return type of LayoutUtil::PaddedDimensions
We should not use const here.
PiperOrigin-RevId:
179614367
A. Unique TensorFlower [Tue, 19 Dec 2017 23:26:33 +0000 (15:26 -0800)]
Remove unused namespace alias.
PiperOrigin-RevId:
179612102
Yao Zhang [Tue, 19 Dec 2017 23:07:16 +0000 (15:07 -0800)]
Support fused batch norm v2.
PiperOrigin-RevId:
179609646
Blake Hechtman [Tue, 19 Dec 2017 23:01:53 +0000 (15:01 -0800)]
[XLA] Turn pad and reduce window of zero sized inputs into a broadcast.
PiperOrigin-RevId:
179608875
Justin Lebar [Tue, 19 Dec 2017 22:56:01 +0000 (14:56 -0800)]
[XLA] Rename "MonotonicDim0MajorLayout" to "DescendingLayout".
This is shorter, and hopefully less confusing.
As a result, we rename NormalizeShapeToMonotonicDim0MajorLayout to the
more descriptive MakeShapeWithDescendingLayoutAndSamePhysicalLayout,
and rename MakeShapewithMonotonicDim0MajorLayout to
MakeShapeWithDescendingLayout.
We also beef up the comments on the first function.
PiperOrigin-RevId:
179608145
A. Unique TensorFlower [Tue, 19 Dec 2017 22:21:51 +0000 (14:21 -0800)]
K-FAC: ResourceVariable support
PiperOrigin-RevId:
179603654
Sukriti Ramesh [Tue, 19 Dec 2017 21:54:53 +0000 (13:54 -0800)]
Migrate SavedModel simple save functionality from contrib to
tensorflow/python/saved_model.
PiperOrigin-RevId:
179599527
A. Unique TensorFlower [Tue, 19 Dec 2017 21:34:33 +0000 (13:34 -0800)]
Fixed a couple of bugs in transitive reduction:
1. Only consider paths through nodes reachable from the source in the inner loop.
2. Ignore paths through Merge nodes, since we cannot know which branch executed.
3. Ignore function nodes, since we are potentially mixing paths from multiple instantiations.
4. Process control inputs to be removed in reverse order of input slot, so we don't clobber the repeated field.
PiperOrigin-RevId:
179596846
Sergio Guadarrama [Tue, 19 Dec 2017 21:26:56 +0000 (13:26 -0800)]
Make sure function.defun can handle nested inputs with namedtuples and dicts.
PiperOrigin-RevId:
179595879
A. Unique TensorFlower [Tue, 19 Dec 2017 20:54:01 +0000 (12:54 -0800)]
Initial checkin for outside_compilation. Adds a new attribute for encapsulating XLA subgraphs that will in the future be used to mark some Ops in the subgraph as 'outside_compilation' meaning they will be run as interpreted TensorFlow via a callout from a compiled XLA subgraph.
This is the first of a sequence of checkins. It adds new types of edges entering and leaving the subgraphs, suitable for send/recv between a compiled XLA subgraph and the 'host', i.e., uncompiled TensorFlow.
For now no code sets the new 'outside_compilation' attributes, and the Ops to perform the send/recv are not present in the codebase; these will follow in subsequent checkins.
PiperOrigin-RevId:
179591853
Yao Zhang [Tue, 19 Dec 2017 20:48:55 +0000 (12:48 -0800)]
Support IdentityN op.
PiperOrigin-RevId:
179591229
A. Unique TensorFlower [Tue, 19 Dec 2017 20:38:19 +0000 (12:38 -0800)]
Protect all calls to launch cuSolver & cuBlas kernels by a lock. The code appears not to be threadsafe pre Cuda 9, and we have several report of crashes. Since the overhead is modest, better to be safe.
PiperOrigin-RevId:
179589983
Austin Anderson [Tue, 19 Dec 2017 20:16:57 +0000 (12:16 -0800)]
Specifically add TMP directory for Android
There is an oversight in the adb tool that causes `adb shell` ->
$<command> to set the TMPDIR env param, but `adb shell <command>` does
not. Binaries that use env.cc on Android will have inconsistent behavior
between interactive and non-interactive versions of `adb shell` until
that bug (which has been fixed internally) arrives on released versions
of adb.
This change adds the /data/local/tmp directory as a final fallback for
Android devices that support it (/data/local/tmp seems to be the de
facto "anyone can use this" Android tmp directory based on Mobile
Harness's use of it, my own discovery that it works better than other
directories, and that searching for it seems to back up the idea, but I
can't find official documentation supporting this). Devices that don't
support that directory will continue to fail, as they did before.
PiperOrigin-RevId:
179587386
A. Unique TensorFlower [Tue, 19 Dec 2017 20:09:52 +0000 (12:09 -0800)]
Support for strings input/output in Toco
PiperOrigin-RevId:
179586555
Shivani Agrawal [Tue, 19 Dec 2017 19:47:21 +0000 (11:47 -0800)]
[tf.data] Saveable iterator for ScanDataset.
PiperOrigin-RevId:
179583460
Shivani Agrawal [Tue, 19 Dec 2017 19:47:16 +0000 (11:47 -0800)]
[tf.data] Saveable iterator for DenseToSparseBatchDataset.
PiperOrigin-RevId:
179583450
Yao Zhang [Tue, 19 Dec 2017 19:46:20 +0000 (11:46 -0800)]
Support more binary ops.
PiperOrigin-RevId:
179583326
A. Unique TensorFlower [Tue, 19 Dec 2017 19:33:15 +0000 (11:33 -0800)]
Remove a series of allocations and memcpys from the GCS read path (TF infeed path).
The GCS client uses an intermediate std::vector<char> buffer for all reads. This turns out to have a substantial cost, in allocations and in memcpys. During load testing with a simple tool for measuring GCS throughput (but not running TensorFlow), removing this allocation and memcpy increased throughput by about 8%, and significantly reduced steady-state heap memory used. I have not measured the impact when running the full TensorFlow pipeline, but I expect there to be gains, due to the lower memory utilization and the lower cache demand.
PiperOrigin-RevId:
179581563
Guangda Lai [Tue, 19 Dec 2017 19:14:38 +0000 (11:14 -0800)]
Fix comment.
PiperOrigin-RevId:
179578952
A. Unique TensorFlower [Tue, 19 Dec 2017 18:11:58 +0000 (10:11 -0800)]
[XLA] Add support for atan2 on CPU
This leans on the libm's atan2 for the actual routine but allows us to share
the implementation of other complex operations between CPU and GPU.
PiperOrigin-RevId:
179569666
HyoukJoong Lee [Tue, 19 Dec 2017 15:43:14 +0000 (07:43 -0800)]
Automated g4 rollback of changelist
179506342
PiperOrigin-RevId:
179552496
A. Unique TensorFlower [Tue, 19 Dec 2017 15:40:36 +0000 (07:40 -0800)]
Run ResizeArea tests on random data rather than all 0s.
PiperOrigin-RevId:
179552297
Malcolm Reynolds [Tue, 19 Dec 2017 14:36:25 +0000 (06:36 -0800)]
Automated g4 rollback of changelist
179529787
PiperOrigin-RevId:
179546698
A. Unique TensorFlower [Tue, 19 Dec 2017 13:48:13 +0000 (05:48 -0800)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
179542713
A. Unique TensorFlower [Tue, 19 Dec 2017 13:36:22 +0000 (05:36 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
179542009
A. Unique TensorFlower [Tue, 19 Dec 2017 10:55:10 +0000 (02:55 -0800)]
Avoid casting to/from int64 in Categorical._sample_n
PiperOrigin-RevId:
179529787
A. Unique TensorFlower [Tue, 19 Dec 2017 07:07:13 +0000 (23:07 -0800)]
[XLA] Adding more tests for conditional.
PiperOrigin-RevId:
179514274
HyoukJoong Lee [Tue, 19 Dec 2017 04:48:26 +0000 (20:48 -0800)]
Changed the buffer liveness such that the buffers pointed to by the root
instruction have liveness to the end of the computation.
PiperOrigin-RevId:
179506342
Guangda Lai [Tue, 19 Dec 2017 04:16:08 +0000 (20:16 -0800)]
Added virtual gpu support.
PiperOrigin-RevId:
179504116
A. Unique TensorFlower [Tue, 19 Dec 2017 03:07:23 +0000 (19:07 -0800)]
[XLA] Hlo parser: don't crash if no computation is specified as ENTRY. Also disallow multiple entry computations and multiple root intructions.
PiperOrigin-RevId:
179498839
Yao Zhang [Tue, 19 Dec 2017 02:53:35 +0000 (18:53 -0800)]
Support more unary ops.
PiperOrigin-RevId:
179498141
Yao Zhang [Tue, 19 Dec 2017 01:45:50 +0000 (17:45 -0800)]
Support more activation ops.
PiperOrigin-RevId:
179493169
Dustin Tran [Tue, 19 Dec 2017 01:15:01 +0000 (17:15 -0800)]
Add probabilistic convolutional layers.
PiperOrigin-RevId:
179490700
Justin Lebar [Tue, 19 Dec 2017 00:43:08 +0000 (16:43 -0800)]
[XLA] Add pow(pow(A, X), Y) => pow(A, X*Y) simplification.
PiperOrigin-RevId:
179487434
A. Unique TensorFlower [Tue, 19 Dec 2017 00:37:40 +0000 (16:37 -0800)]
Fix typo in comment for GetRoot().
PiperOrigin-RevId:
179486882
A. Unique TensorFlower [Mon, 18 Dec 2017 23:14:59 +0000 (15:14 -0800)]
[XLA] Add format field to layout
Format will describe the method used to store array data in memory. Currently
only DENSE is supported, which represents the way XLA currently stores arrays.
Scalars have a DENSE format. Tuples and opaque shapes use INVALID_FORMAT.
Adds checks to code that uses minor_to_major to ensure the layout is dense.
PiperOrigin-RevId:
179475450
Jonathan Hseu [Mon, 18 Dec 2017 22:01:58 +0000 (14:01 -0800)]
Delete the instructions regarding manually installing protobuf.
PiperOrigin-RevId:
179464468
Sukriti Ramesh [Mon, 18 Dec 2017 21:20:19 +0000 (13:20 -0800)]
SavedModel tag logging.
PiperOrigin-RevId:
179460064