| 5c9e7dc293cbeda6b085238fabacba02560424b4 |
|
23-Oct-2017 |
Yong Tang <yong.tang.github@outlook.com> |
Sanitize files in tensorflow/core/ops with clang-format -i
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| 0fe0bfcc3cf6930edc096998b1445cead92de8c3 |
|
06-Jun-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Remove unused protobuf header inclusions
PiperOrigin-RevId: 158120864
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| eb10a4c494d95e7c17ddc44ef35197d08f2f6b33 |
|
01-Jun-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Preallocate vector storage when the ultimate vector size is known in advance
PiperOrigin-RevId: 157724431
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| be786946f7a0db68977fa95bfb77fbe7e9a8265d |
|
28-Oct-2016 |
Eugene Brevdo <ebrevdo@google.com> |
Add improved shape information to sparse_tensor_dense_matmul.
Change: 137520986
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| fdea17d8b449cbee9719ab4022a24e2d9918c25f |
|
11-Oct-2016 |
Eugene Brevdo <ebrevdo@google.com> |
Store SparseTensors in a Map inside a container for Queue round-trip.
This is much more efficient than serializing the underlying Tensors to strings
and dserializing them on the other side. Instead we pass through the keys
to the SparseTensors inside the Map.
Methods are kept private for use by queueing wrappers.
Includes benchmarks that show wall-time is almost 50% of the wall-time of using the sparse serialization/deserialization wrappers:
I1003 17:24:34.355306 18675 benchmark.py:77] Benchmark [BenchmarkSparseTensorsMapVsSerialization.benchmark_very_large_2d_float_st_tensor_maps] iters: 2000, wall_time: 0.00260997, cpu_time: -1,throughput: -1
I1003 17:24:42.735983 18675 benchmark.py:77] Benchmark [BenchmarkSparseTensorsMapVsSerialization.benchmark_vey_large_2d_float_st_serialization] iters: 2000, wall_time: 0.00415492, cpu_time: -1,throughput: -1
*** Update:
After updates to sparse_tensor.h's concat code (pushed in a sister PR), there's a speedup in both benchmarks:
I1004 09:39:30.630354 24400 benchmark.py:77] Benchmark [BenchmarkSparseTensorsMapVsSerialization.benchmark_very_large_2d_float_st_tensor_maps] iters: 2000, wall_time: 0.0022105
I1004 09:39:38.125391 24400 benchmark.py:77] Benchmark [BenchmarkSparseTensorsMapVsSerialization.benchmark_very_large_2d_float_st_serialization] iters: 2000, wall_time: 0.00372696
*** Update 2:
After properly placed std::moves in the sparse_tensors_map code, that benchmark is now faster:
Benchmark [BenchmarkSparseTensorsMapVsSerialization.benchmark_very_large_2d_float_st_tensor_maps] iters: 2000, wall_time: 0.00187492
Total speedup is now: 0.00415492 / 0.00187492 = 2.2x
Change: 135805924
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| 92052d14875ae0ab3daeacabcaa24e239d50f193 |
|
02-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Enable C++ shape function for math_ops.py and array_ops.cc shape functions that
don't require access to the input tensor values.
Change: 132071962
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| e11b99749d2898df0bce0269c77df316b059e8ea |
|
01-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Automated rollback of change 131739513
Change: 131887027
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| a7fc9f53fa928e85b5d2b037efdc0e84f7a7b31d |
|
30-Aug-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Enable C++ shape function for math_ops.py and array_ops.cc shape functions that
don't require access to the input constant tensor, and for which the C++
function's behavior matches python.
Fix rank check in C++ BatchMatMul shape function.
Fix rank check in C++ sparse cwise shape function.
Add an option on common_shapes.call_cpp_shape_fn to verify against the python
shape function. This is meant to be called only when preparing a change to
switch shape functions over.
Change: 131739513
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| cf93b043a88607db17b84b5b5d3f0064473cf5ce |
|
12-Aug-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add C++ shape inference functions for ParseExample and
ParseSingleSequenceExample.
The attribute parsing code was moved over verbatim to example_proto_helper.*.
Change TF_CHECK_OK to TF_ASSERT_OK in all shape inference tests under core.
Change: 130041336
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|
| f3b61805997f2a08fab35bd02f0f6be18e6cd8cc |
|
22-Jul-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Translate shape inference functions for sparse_ops to C++.
Change shape_inference::InferenceContext::MakeShapeFromShapetensor to handle
the case where the shape tensor is not known but the rank of the shape tensor
is.
Add shape_inference::NumElements().
Change: 128124939
/external/tensorflow/tensorflow/core/ops/sparse_ops_test.cc
|