A. Unique TensorFlower [Wed, 18 Apr 2018 18:46:07 +0000 (11:46 -0700)]
Change operands of subtraction expression to have well-defined behaviour.
At present, signed arithmetic overflows (i.e. has undefined behaviour) in general, e.g. when computing 0 - INT_MIN or INT_MAX - INT_MIN. The fact that we want the result in the unsigned type does not help us here.
The fix is to convert the operands to the corresponding unsigned type first and then perform the operation in unsigned arithmetic, which is well-defined and has the correct subtraction behaviour.
PiperOrigin-RevId:
193391813
A. Unique TensorFlower [Wed, 18 Apr 2018 17:28:47 +0000 (10:28 -0700)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
193378249
A. Unique TensorFlower [Wed, 18 Apr 2018 17:27:48 +0000 (10:27 -0700)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
193378087
A. Unique TensorFlower [Wed, 18 Apr 2018 16:40:21 +0000 (09:40 -0700)]
Simplify the break canonicalization transformer to use more of the base transformer helpers. Add support for the loop's else block.
PiperOrigin-RevId:
193370640
A. Unique TensorFlower [Wed, 18 Apr 2018 16:30:17 +0000 (09:30 -0700)]
Updating some more tests in constant_folding_test.cc so that the tests evaluate the original and optimized graphs and check if their outputs are the same.
PiperOrigin-RevId:
193369280
Sanjoy Das [Wed, 18 Apr 2018 16:03:21 +0000 (09:03 -0700)]
Avoid generating degenerate dimensions during gather expansions
This gets rid of two cases that used to introduce degenerate dimensions
(dimensions with bound = 1) into the while loop state:
- Previously we'd explicitly reshape gathers using scalar indices to have a
minor degenerate dimension. With this CL we no longer do that - instead we
push this into the code that looks up the index vector from the gather
indices tensor.
- Previously we'd have the accumulator (the tensor we're
dynamic-update-slice-ing into) contain all of the degenerate window dims that
the gather op would later elide (after the while loop). With this CL we
eagerly elide these dimensions as we slice out individual windows from the
operand.
PiperOrigin-RevId:
193365863
A. Unique TensorFlower [Wed, 18 Apr 2018 15:40:40 +0000 (08:40 -0700)]
Allow default min/max ranges for int16 data types.
PiperOrigin-RevId:
193362891
Skye Wanderman-Milne [Wed, 18 Apr 2018 15:10:05 +0000 (08:10 -0700)]
Fix bug in importing while loops within a while loop.
PiperOrigin-RevId:
193358699
Derek Murray [Wed, 18 Apr 2018 07:18:50 +0000 (00:18 -0700)]
Enable the n=1 special case in the DeserializeSparse op.
The optimized case was previously dead because of two off-by-one errors (mea culpa).
PiperOrigin-RevId:
193314065
Noah Eisen [Wed, 18 Apr 2018 06:06:57 +0000 (23:06 -0700)]
No public changes.
PiperOrigin-RevId:
193309262
Justin Lebar [Wed, 18 Apr 2018 04:04:35 +0000 (21:04 -0700)]
[XLA] Convert XLA to use xla::se as a namespace alias for ::stream_executor.
PiperOrigin-RevId:
193301997
Chris Ying [Wed, 18 Apr 2018 03:31:30 +0000 (20:31 -0700)]
Disable CheckpointSaverHook when both save_checkpoints_secs and
save_checkpoints_steps are None
PiperOrigin-RevId:
193299688
A. Unique TensorFlower [Wed, 18 Apr 2018 01:48:38 +0000 (18:48 -0700)]
Comment out part of ring_reducer_test suspected to have a race.
PiperOrigin-RevId:
193292788
Sanjoy Das [Wed, 18 Apr 2018 00:17:04 +0000 (17:17 -0700)]
Minor cleanups to the gather expander; NFC
This change is NFC now, but it makes the code more general and this generality
will be used later on. For instance
ExpandFirstDimIntoNDims(transposed_gather_indices, {1, shape.dimensions(0)})
does not work if shape is a scalar shape (and this fine because today shape is
never scalar) but
PrependDegenerateDims(transposed_gather_indices, 1)
works fine if transposed_gather_indices is scalar (and it will be, in a future
change).
PiperOrigin-RevId:
193283404
Jonathan Hseu [Tue, 17 Apr 2018 23:24:42 +0000 (16:24 -0700)]
TpuStrategy -> TPUStrategy
PiperOrigin-RevId:
193275991
Chris Ying [Tue, 17 Apr 2018 22:28:12 +0000 (15:28 -0700)]
Fix CheckpointSaverHook to properly save every save_checkpoints_steps for TPU workloads.
PiperOrigin-RevId:
193266515
Michael Case [Tue, 17 Apr 2018 22:01:22 +0000 (15:01 -0700)]
Internal Change.
PiperOrigin-RevId:
193262066
Jacques Pienaar [Tue, 17 Apr 2018 21:47:44 +0000 (14:47 -0700)]
Don't consider control flow edges when computing switch depth of switch.
PiperOrigin-RevId:
193259710
A. Unique TensorFlower [Tue, 17 Apr 2018 21:33:53 +0000 (14:33 -0700)]
Use easy_install to update pip only on Ubuntu14.04
* We only depends on easy_install/easy_install3 to update pip on Ubuntu14.04
* They are not always available for later system e.g. debian9
* We can use pip/pip3 to update themselves
PiperOrigin-RevId:
193257326
Justin Lebar [Tue, 17 Apr 2018 21:26:16 +0000 (14:26 -0700)]
[StreamExecutor] Rename ::perftools::gputools -> ::stream_executor, part 1.
Step 1 of re-namespace'ing StreamExecutor into ::stream_executor.
This moves everything inside of stream_executor/..., and leaves a
namespace alias into ::perftools::gputools. The next steps will clean
up users to use the new namespace.
This is mostly a mechanical change, but it also includes a bunch of
non-mechanical changes that ideally would be split out into separate
patches. Unfortunately they all sort of need to be shoved in here for
various reasons:
- forward declarations need to be in the same namespace as the actual
types, so we need to change all forward declarations of
StreamExecutor types in this one patch.
- Uses of these forward declarations need to be changed to the new
namespace (or otherwise we need to add a namespace alias to the
relevant header, but this is pretty ugly).
- Various initialization code needs to live in StreamExecutor's "real"
namespace, so all this needs to be changed.
PiperOrigin-RevId:
193256128
A. Unique TensorFlower [Tue, 17 Apr 2018 20:36:46 +0000 (13:36 -0700)]
[XLA] Redesign: support xla::XlaComputation in compile-only client and service.
PiperOrigin-RevId:
193247845
A. Unique TensorFlower [Tue, 17 Apr 2018 20:36:24 +0000 (13:36 -0700)]
Make GroupRec* const in GroupRecCallback by marking mu mutable in
CollectiveParamResolverLocal::GroupRec.
PiperOrigin-RevId:
193247799
Akshay Modi [Tue, 17 Apr 2018 20:31:54 +0000 (13:31 -0700)]
Use is_constructible instead of is_convertible.
Before this, all objects would follow the slow path (since nothing is
convertible to AlphaNum since it has a private copy constructor).
Before:
entry {
name: "MicroBenchmarks.benchmark_defun_matmul_2_by_2_CPU"
iters: 30000
wall_time: 67.
5895690918
extras {
key: "examples_per_sec"
value {
double_value: 14795.1823549
}
}
}
After:
entry {
name: "MicroBenchmarks.benchmark_defun_matmul_2_by_2_CPU"
iters: 30000
wall_time: 61.
0044002533
extras {
key: "examples_per_sec"
value {
double_value: 16392.2601623
}
}
}
PiperOrigin-RevId:
193247183
A. Unique TensorFlower [Tue, 17 Apr 2018 20:30:45 +0000 (13:30 -0700)]
Unpack multiple assignments when processing lists. This supports the cases "a, b = [], []" and "[a, b] = [], []". Also expand the static analysis to support list unpacking constructs.
PiperOrigin-RevId:
193247024
Allen Lavoie [Tue, 17 Apr 2018 20:27:46 +0000 (13:27 -0700)]
Internal change.
PiperOrigin-RevId:
193246563
A. Unique TensorFlower [Tue, 17 Apr 2018 20:20:42 +0000 (13:20 -0700)]
Fixing a bug in strided slice. The op was not handling negative indices correctly.
PiperOrigin-RevId:
193245539
A. Unique TensorFlower [Tue, 17 Apr 2018 19:52:34 +0000 (12:52 -0700)]
Cache the ag_internal module, to avoid falsely rejecting it when in the namespace of a previously converted function. Explicitly reject lambda functions, for now, becasue they require special treatment.
PiperOrigin-RevId:
193241279
A. Unique TensorFlower [Tue, 17 Apr 2018 19:36:20 +0000 (12:36 -0700)]
[XLA] Fix arguments to IsSparseArray and IsDenseArray
PiperOrigin-RevId:
193238920
A. Unique TensorFlower [Tue, 17 Apr 2018 19:31:42 +0000 (12:31 -0700)]
Fix the test gensym to prevent creating duplicate names in the same test.
PiperOrigin-RevId:
193238314
A. Unique TensorFlower [Tue, 17 Apr 2018 19:31:36 +0000 (12:31 -0700)]
Relaxes the type constraints for the features in ServingInputReceiver, so it will accept anything convertible to a Tensor or SparseTensor.
This makes it possible to use with tf.contrib.labeled_tensor.
PiperOrigin-RevId:
193238295
Noah Eisen [Tue, 17 Apr 2018 19:21:10 +0000 (12:21 -0700)]
Removes another custom implementation of ZeroCopyInputStream, instead uses the now public gRPC implementation.
Also, moves GrpcByteSource to grpc_util, to keep it near the other serialization code.
Lastly, gives a more verbose error if serialization (unparsing) fails (which should not ever happen).
PiperOrigin-RevId:
193236893
Yifei Feng [Tue, 17 Apr 2018 19:07:33 +0000 (12:07 -0700)]
Automated g4 rollback of changelist
193168327
PiperOrigin-RevId:
193234819
Pavithra Vijay [Tue, 17 Apr 2018 19:06:50 +0000 (12:06 -0700)]
Moving gradient registration for CudnnRNN op from contrib to core.
PiperOrigin-RevId:
193234663
A. Unique TensorFlower [Tue, 17 Apr 2018 18:53:29 +0000 (11:53 -0700)]
Fixes and cleanup to support more complex quantized models and adds PropagateFakeQuantNumBits.
PiperOrigin-RevId:
193232630
Mingxing Tan [Tue, 17 Apr 2018 18:48:43 +0000 (11:48 -0700)]
Replace decode_image with decode_jpeg to avoid ValueError in datasets programmers guide.
PiperOrigin-RevId:
193231717
A. Unique TensorFlower [Tue, 17 Apr 2018 18:05:57 +0000 (11:05 -0700)]
Fixes a comment in tf.contrib.seq2seq.monotonic_attention().
PiperOrigin-RevId:
193224285
Akshay Modi [Tue, 17 Apr 2018 17:32:47 +0000 (10:32 -0700)]
Avoid ToString() in Eager's TFE_Execute.
Also use InlinedVector instead of std::vector for non-async path
Before:
Benchmark Time(ns) CPU(ns) Iterations
-------------------------------------------------------------
BM_Execute/0 1895 1898 360200 Execute
BM_Execute/1 1193 1942 358322 ExecuteAsync
BM_ExecuteFunction/0 5812 5825 100000 ExecuteFunction
BM_ExecuteFunction/1 5015 5374 100000 ExecuteFunctionAsync
After:
Benchmark Time(ns) CPU(ns) Iterations
-------------------------------------------------------------
BM_Execute/0 1604 1607 428262 Execute
BM_Execute/1 1150 1765 404821 ExecuteAsync
BM_ExecuteFunction/0 5615 5626 100000 ExecuteFunction
BM_ExecuteFunction/1 5111 5476 100000 ExecuteFunctionAsync
PiperOrigin-RevId:
193218331
A. Unique TensorFlower [Tue, 17 Apr 2018 17:13:09 +0000 (10:13 -0700)]
Change the contract of dynamic_builtin to reject all functions it can't process.
PiperOrigin-RevId:
193215246
Max Galkin [Tue, 17 Apr 2018 16:19:06 +0000 (09:19 -0700)]
Estimate IdentityN as Identity.
PiperOrigin-RevId:
193207469
A. Unique TensorFlower [Tue, 17 Apr 2018 16:02:14 +0000 (09:02 -0700)]
Fix incorrect rejection of xrange.
PiperOrigin-RevId:
193205016
A. Unique TensorFlower [Tue, 17 Apr 2018 14:41:09 +0000 (07:41 -0700)]
Add uint32/uint64 support to Gather op.
PiperOrigin-RevId:
193195939
A. Unique TensorFlower [Tue, 17 Apr 2018 14:07:16 +0000 (07:07 -0700)]
Fix the type info analysis to correctly process loops. Simplify the implementation by reusing some of the transformer base functions. Allow set_element_type to use literals. Add additional tests.
PiperOrigin-RevId:
193192409
A. Unique TensorFlower [Tue, 17 Apr 2018 09:23:10 +0000 (02:23 -0700)]
Automated g4 rollback of changelist
192842670
PiperOrigin-RevId:
193168327
Derek Murray [Tue, 17 Apr 2018 06:07:24 +0000 (23:07 -0700)]
[tf.data] Add an API for randomly sampling from multiple datasets.
Fixes #15999.
PiperOrigin-RevId:
193152683
Saurabh Saxena [Tue, 17 Apr 2018 03:28:02 +0000 (20:28 -0700)]
Make ParallelInterleaveDataset saveable.
PiperOrigin-RevId:
193142302
A. Unique TensorFlower [Tue, 17 Apr 2018 02:51:13 +0000 (19:51 -0700)]
Supply a dtype to super constructor, without which build() seems to crash.
PiperOrigin-RevId:
193139585
Younghee Kwon [Tue, 17 Apr 2018 02:10:10 +0000 (19:10 -0700)]
BoostedTreesEstimator in contrib: train_in_memory works with input_fns returning data.Dataset.
Only one batch of data is expected, so dataset.batch() is disallowed,
and dataset.repeat() will be ignored (only the first one would be used)
PiperOrigin-RevId:
193137094
Yifei Feng [Tue, 17 Apr 2018 01:41:28 +0000 (18:41 -0700)]
Remove proto imports in header files for core/kernels/hexagon.
The goal is to make kernels mostly independent of proto headers, which will let us lock down our .so imports.
PiperOrigin-RevId:
193134710
Shashi Shekhar [Tue, 17 Apr 2018 01:31:22 +0000 (18:31 -0700)]
Add a simple Profiler and instrument operator invocations in Interpreter.
PiperOrigin-RevId:
193133955
A. Unique TensorFlower [Tue, 17 Apr 2018 01:03:05 +0000 (18:03 -0700)]
Adding min node weight regularization
PiperOrigin-RevId:
193131300
Michael Case [Tue, 17 Apr 2018 00:34:01 +0000 (17:34 -0700)]
Remove deprecated/unused python related Bazel options.
Since py_runtime was introduced, Bazel ignores options such as
--force_python2 and --python2_path. Deleting to clean stuff up and
make sure people are not misled.
PiperOrigin-RevId:
193127681
Derek Murray [Tue, 17 Apr 2018 00:25:12 +0000 (17:25 -0700)]
[tf.data] Sort the results of `tf.matching_files()` to enable `Dataset.list_files()` to be determinstic.
PiperOrigin-RevId:
193126572
Benoit Steiner [Tue, 17 Apr 2018 00:12:05 +0000 (17:12 -0700)]
Don't rely on graph contruction for an initial shape inference.
PiperOrigin-RevId:
193124836
A. Unique TensorFlower [Tue, 17 Apr 2018 00:05:18 +0000 (17:05 -0700)]
Add LinearOperatorKronecker, representing the Kronecker product.
PiperOrigin-RevId:
193123894
Anna R [Mon, 16 Apr 2018 23:41:12 +0000 (16:41 -0700)]
Internal change.
PiperOrigin-RevId:
193119953
A. Unique TensorFlower [Mon, 16 Apr 2018 23:34:18 +0000 (16:34 -0700)]
Port the list append into the operators module. Not enabled yet.
PiperOrigin-RevId:
193118940
Olivia Nordquist [Mon, 16 Apr 2018 23:24:19 +0000 (16:24 -0700)]
disabling flaky asan test
PiperOrigin-RevId:
193117611
A. Unique TensorFlower [Mon, 16 Apr 2018 23:22:03 +0000 (16:22 -0700)]
Internal change
PiperOrigin-RevId:
193117142
A. Unique TensorFlower [Mon, 16 Apr 2018 23:15:19 +0000 (16:15 -0700)]
Create copy of locals() before copying, since modifying locals does not always affect the values. https://docs.python.org/2/library/functions.html#locals.
PiperOrigin-RevId:
193116254
A. Unique TensorFlower [Mon, 16 Apr 2018 23:04:43 +0000 (16:04 -0700)]
Replacing the current inner Cholesky decomposition loop with a While loop rolled version.
This will allow for much larger Cholesky decompositions (and thus matrix inversions) than previously possible on TPU because of the use of rolled While loops so XLA compilation will no longer timeout.
While there is a minor runtime performance decrease (now 25ms vs 15ms for a 500x500 matrix) the compilation time is significantly faster (12.8s vs 55.2s for a 500x500 matrix.)
PiperOrigin-RevId:
193114816
A. Unique TensorFlower [Mon, 16 Apr 2018 22:52:54 +0000 (15:52 -0700)]
Fix trace collection to properly remove the suffix.
PiperOrigin-RevId:
193113074
A. Unique TensorFlower [Mon, 16 Apr 2018 22:46:42 +0000 (15:46 -0700)]
Internal change.
PiperOrigin-RevId:
193112205
James Qin [Mon, 16 Apr 2018 21:52:41 +0000 (14:52 -0700)]
Increase softmax gpu unittest numeric stability
PiperOrigin-RevId:
193103363
Jiri Simsa [Mon, 16 Apr 2018 21:47:31 +0000 (14:47 -0700)]
Porting tests for `rpc_op` to OS.
PiperOrigin-RevId:
193102564
A. Unique TensorFlower [Mon, 16 Apr 2018 21:41:21 +0000 (14:41 -0700)]
Exposes InputPipelineConfig as tf.contrib.tpu.InputPipelineConfig.
This type is expected by the `per_host_input_for_training` argument of the TPUConfig constructor, but is not currently visible.
PiperOrigin-RevId:
193101540
Olivia Nordquist [Mon, 16 Apr 2018 21:17:02 +0000 (14:17 -0700)]
disabling test that fails tensorflow.asan
PiperOrigin-RevId:
193097794
A. Unique TensorFlower [Mon, 16 Apr 2018 21:13:52 +0000 (14:13 -0700)]
Use fixed sized tensor arrays and max loop iterations in dynamic_decode if the user supplies it and if the inputs were created in an XLA context.
PiperOrigin-RevId:
193097293
A. Unique TensorFlower [Mon, 16 Apr 2018 20:54:03 +0000 (13:54 -0700)]
Simplify the recursion when processing unpackings.
PiperOrigin-RevId:
193094078
Nick Desaulniers [Mon, 16 Apr 2018 20:36:55 +0000 (13:36 -0700)]
[XLA] Document and enforce reduction order of init value
All existing backends apply the init_value on the lhs, except for the evaluator. This causes problems for reductions which apply an identity function to a reduce or reduce window.
PiperOrigin-RevId:
193091323
A. Unique TensorFlower [Mon, 16 Apr 2018 20:32:12 +0000 (13:32 -0700)]
Adding several utility functions to TF2XLA to help with the Cholesky refactor. Mainly responsible for handling batching properly.
PiperOrigin-RevId:
193090634
A. Unique TensorFlower [Mon, 16 Apr 2018 20:22:58 +0000 (13:22 -0700)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
193089301
A. Unique TensorFlower [Mon, 16 Apr 2018 20:04:23 +0000 (13:04 -0700)]
Simplify the implementation of break_statements.py
PiperOrigin-RevId:
193086371
Younghee Kwon [Mon, 16 Apr 2018 19:56:14 +0000 (12:56 -0700)]
boosted_trees: Make some regularizer/hyper-params as inputs instead of attributes.
PiperOrigin-RevId:
193085059
Igor Saprykin [Mon, 16 Apr 2018 19:21:15 +0000 (12:21 -0700)]
Early TPU distribution strategy and the associated testing infrastructure.
PiperOrigin-RevId:
193080098
A. Unique TensorFlower [Mon, 16 Apr 2018 19:12:46 +0000 (12:12 -0700)]
Tighten label check in BinaryLogisticHeadWithSigmoidCrossEntropyLoss
PiperOrigin-RevId:
193078844
A. Unique TensorFlower [Mon, 16 Apr 2018 19:09:36 +0000 (12:09 -0700)]
Add a common transformer feature that allows keeping temporary state across nodes. To be used in the break, continue and return canonicalizers.
PiperOrigin-RevId:
193078359
A. Unique TensorFlower [Mon, 16 Apr 2018 19:09:33 +0000 (12:09 -0700)]
Copy the if statement handlers over to the operators module. They will enabled in a follow-up CL.
PiperOrigin-RevId:
193078348
A. Unique TensorFlower [Mon, 16 Apr 2018 19:09:24 +0000 (12:09 -0700)]
Remove obsolete tests. Patch the unexpected print output in Python 3.
PiperOrigin-RevId:
193078330
A. Unique TensorFlower [Mon, 16 Apr 2018 18:56:15 +0000 (11:56 -0700)]
Mark the parent list as modified for index writes. Add special case for constructors where we know setting an attribute actually creates a new symbol. Clean up the tests a bit.
PiperOrigin-RevId:
193075909
A. Unique TensorFlower [Mon, 16 Apr 2018 18:47:11 +0000 (11:47 -0700)]
Refactoring: Rename the __ops module to ag__ (double underscore prefix has special meaning in Python). Consolidate all internal API calls through the ag__ module.
PiperOrigin-RevId:
193074379
Austin Anderson [Mon, 16 Apr 2018 18:33:14 +0000 (11:33 -0700)]
Internal testing changes
PiperOrigin-RevId:
193071881
Igor Saprykin [Mon, 16 Apr 2018 18:27:09 +0000 (11:27 -0700)]
Remove the hidden replicate_model_fn copy from core.
PiperOrigin-RevId:
193070799
Jiri Simsa [Mon, 16 Apr 2018 18:24:43 +0000 (11:24 -0700)]
Porting tests for the `decode_proto` and `encode_proto` to OS.
PiperOrigin-RevId:
193070420
Anna R [Mon, 16 Apr 2018 18:20:40 +0000 (11:20 -0700)]
Prefix clip_by_value with underscore in gen_math_ops so that it doesn't
interfere with clip_ops.clip_by_value when importing.
PiperOrigin-RevId:
193069700
Anna R [Mon, 16 Apr 2018 17:22:29 +0000 (10:22 -0700)]
PiperOrigin-RevId:
193059174
Benjamin Kramer [Mon, 16 Apr 2018 16:34:27 +0000 (09:34 -0700)]
Update for ObjectMemoryBuffer.h rename in upstream LLVM.
This will require a version bump in workspace.bzl
PiperOrigin-RevId:
193052084
A. Unique TensorFlower [Mon, 16 Apr 2018 04:52:17 +0000 (21:52 -0700)]
Internal change
PiperOrigin-RevId:
192981122
A. Unique TensorFlower [Mon, 16 Apr 2018 00:35:39 +0000 (17:35 -0700)]
Minor cleanup.
PiperOrigin-RevId:
192971080
A. Unique TensorFlower [Sat, 14 Apr 2018 09:15:58 +0000 (02:15 -0700)]
Style nit: avoid creating local variables when not required.
PiperOrigin-RevId:
192876802
A. Unique TensorFlower [Sat, 14 Apr 2018 08:22:59 +0000 (01:22 -0700)]
Make sure that same nodes are not optimized as part of multiple groups.
Replace recusrsion with iteration in AbsorbInputByOptimizedNodesGroup.
PiperOrigin-RevId:
192874364
A. Unique TensorFlower [Sat, 14 Apr 2018 04:15:59 +0000 (21:15 -0700)]
[XLA] Use pattern matcher in algebraic simplifier
PiperOrigin-RevId:
192862841
A. Unique TensorFlower [Sat, 14 Apr 2018 04:09:37 +0000 (21:09 -0700)]
Upgrade gRPC version used in OSS Tensorflow
PiperOrigin-RevId:
192862541
A. Unique TensorFlower [Sat, 14 Apr 2018 04:00:02 +0000 (21:00 -0700)]
Check there are no duplicate entries in sparse features as this would invalidate the example norm computation in SDCA.
PiperOrigin-RevId:
192861834
A. Unique TensorFlower [Sat, 14 Apr 2018 03:59:14 +0000 (20:59 -0700)]
Adding 1d and 3d orthogonal kernels convolutions.
PiperOrigin-RevId:
192861809
A. Unique TensorFlower [Sat, 14 Apr 2018 02:22:02 +0000 (19:22 -0700)]
Internal Change
PiperOrigin-RevId:
192856330
A. Unique TensorFlower [Sat, 14 Apr 2018 02:18:28 +0000 (19:18 -0700)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
192856167
Igor Saprykin [Sat, 14 Apr 2018 02:09:07 +0000 (19:09 -0700)]
Clarify a caveat about metrics.
PiperOrigin-RevId:
192855733
Scott Zhu [Sat, 14 Apr 2018 00:52:20 +0000 (17:52 -0700)]
Merge changes from github.
PiperOrigin-RevId:
192850372
Asim Shankar [Sat, 14 Apr 2018 00:51:37 +0000 (17:51 -0700)]
Java: Bump release to 1.8.0-rc0
PiperOrigin-RevId:
192850310
Igor Saprykin [Sat, 14 Apr 2018 00:47:58 +0000 (17:47 -0700)]
Add a caveat about make_initiliazable_iterator to the README.
PiperOrigin-RevId:
192850014