| .. |
|
batching_util
|
Change queues in SharedBatchScheduler to have different traceme_context_id_counter_.
|
2021-02-02 13:57:29 -08:00 |
|
boosted_trees
|
Comment out a number of google-internal targets when copybara-exporting instead of removing them.
|
2021-01-26 05:47:31 -08:00 |
|
data
|
Add space for reporting memory bandwidth statistics
|
2021-02-09 00:08:28 -08:00 |
|
fuzzing
|
Lower memory threshold for parse_tensor_op_fuzz
|
2020-11-20 12:18:27 -08:00 |
|
hexagon
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
image
|
Disable two gpu test breaking by cl/355688576 at the moment.
|
2021-02-04 20:36:23 -08:00 |
|
linalg
|
Add tf.MatrixDeterminant ops to Flex
|
2021-02-02 15:20:06 -08:00 |
|
mkl
|
Merge pull request #45817 from Intel-tensorflow:dnn0x_clean_matmul_fused_2
|
2021-01-07 22:41:57 -08:00 |
|
mlir_generated
|
[mlir][kernelgen] Test ops that can return NaN with the default set of inputs
|
2021-02-09 04:42:16 -08:00 |
|
neon
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
risc
|
Expand risc ops and update shape function for binary ops.
|
2021-01-22 15:56:21 -08:00 |
|
rnn
|
Remove depencencies on unnecessary aliases in tensorflow/core
|
2020-10-13 10:28:00 -07:00 |
|
sparse
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
special_math
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
spectrogram_test_data
|
Create BUILD files and corresponding targets for tensorflow/core/kernels/spectrogram_test_data/BUILD.
|
2020-11-03 16:09:26 -08:00 |
|
tensor_forest
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
aggregate_ops_cpu.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
aggregate_ops_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
aggregate_ops.cc
|
Remove OpKernel* templating from AddNOp.
|
2020-12-15 18:36:46 -08:00 |
|
aggregate_ops.h
|
Remove OpKernel* templating from AddNOp.
|
2020-12-15 18:36:46 -08:00 |
|
argmax_op_gpu.cu.cc
|
Support argmin/argmax of boolean tensors on CPUs and GPUs.
|
2020-04-14 13:37:21 -07:00 |
|
argmax_op.cc
|
Support argmin/argmax of boolean tensors on CPUs and GPUs.
|
2020-04-14 13:37:21 -07:00 |
|
argmax_op.h
|
|
|
|
as_string_op_test.cc
|
AddN for variants adds in a tree structure (pairwise summation)
|
2020-11-02 09:11:55 -08:00 |
|
as_string_op.cc
|
AddN for variants adds in a tree structure (pairwise summation)
|
2020-11-02 09:11:55 -08:00 |
|
assign_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
avgpooling_op_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
avgpooling_op.cc
|
Rollforward of 93c22a269b. Windows and RCOM builds have been fixed.
|
2020-08-31 14:33:25 -07:00 |
|
avgpooling_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
barrier_ops.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
base64_ops.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
basic_ops_benchmark_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
batch_kernels.cc
|
Fix potential mismatched function library runtime/handle.
|
2021-01-21 23:23:56 -08:00 |
|
batch_norm_op_gpu.cu.cc
|
|
|
|
batch_norm_op_test.cc
|
|
|
|
batch_norm_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
batch_norm_op.h
|
|
|
|
batch_util.h
|
batch_util.h is generally useful so moved to util/ from kernels/ where it will be included in the pip package.
|
2018-05-22 02:24:26 -07:00 |
|
batchtospace_op.cc
|
|
|
|
bcast_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
betainc_op_gpu.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
betainc_op.cc
|
|
|
|
betainc_op.h
|
|
|
|
bias_op_gpu.cu.cc
|
Add int32 type support for BiasAdd. CPU support already exists, so adding GPU support only.
|
2019-10-31 15:58:08 -07:00 |
|
bias_op_gpu.h
|
ENH: The gradient op of bias_add supports 3/4/5D NCHW format
|
2018-12-28 11:47:13 +08:00 |
|
bias_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
bias_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
bias_op.h
|
Remove redundant reshape from BiasOp
|
2019-10-17 21:35:51 -07:00 |
|
bincount_op_gpu.cu.cc
|
Merge pull request #42431 from GHGmc2:master
|
2020-08-26 16:56:43 -07:00 |
|
bincount_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
bincount_op.cc
|
Fix tf.bincount errors in eager mode
|
2020-07-28 21:56:14 -07:00 |
|
bincount_op.h
|
Add bincount to support dense/sparse/ragged inputs.
|
2020-05-06 14:00:20 -07:00 |
|
broadcast_to_op_gpu.cu.cc
|
|
|
|
broadcast_to_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
broadcast_to_op.cc
|
Register int64 kernels for BroadcastTo
|
2020-04-28 12:19:44 -07:00 |
|
broadcast_to_op.h
|
Implement tf.broadcast_to in a way the can handle higher ranks. This version uses the broadcast optimizations in core/util/bcast.h, which is used for most broadcasting ops in TensorFlow.
|
2019-02-07 17:29:26 -08:00 |
|
bucketize_op_gpu.cu.cc
|
added __restrict__ to all possible __global__ kernels
|
2019-08-01 15:25:07 +09:00 |
|
bucketize_op.cc
|
Enable GPU tests for bucketize op and fix a hard crash on empty input tensors.
|
2019-09-09 20:49:25 -07:00 |
|
bucketize_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
BUILD
|
Replace MKL-DNN v0.x in vanilla TensorFlow and TFRT with DNNL/oneDNN v1.8.
|
2021-02-05 11:35:14 -08:00 |
|
candidate_sampler_ops.cc
|
Remove (Mutable)ArraySlice implementation and alias them to absl::Span.
|
2018-08-30 11:29:03 -07:00 |
|
cast_op_gpu.cu.cc
|
|
|
|
cast_op_impl_bfloat.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
cast_op_impl_bool.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_complex64.cc
|
|
|
|
cast_op_impl_complex128.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
cast_op_impl_double.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_float.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_half.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
cast_op_impl_int8.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_int16.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_int32.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_int64.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_uint8.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_uint16.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_uint32.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl_uint64.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_impl.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
cast_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cast_op.h
|
Merge pull request #40962 from Intel-tensorflow:yang/eigen-bf16
|
2020-07-14 17:21:38 -07:00 |
|
check_numerics_op_gpu.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
check_numerics_op.cc
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
clustering_ops_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
clustering_ops.cc
|
|
|
|
collective_nccl_broadcaster.cc
|
Make NcclManager part of CollectiveExecutorMgr
|
2020-09-17 14:35:16 -07:00 |
|
collective_nccl_broadcaster.h
|
PR #31485: [ROCm] add ROCm RCCL support
|
2019-08-22 13:08:13 -07:00 |
|
collective_nccl_gatherer.cc
|
Make NcclManager part of CollectiveExecutorMgr
|
2020-09-17 14:35:16 -07:00 |
|
collective_nccl_gatherer.h
|
|
|
|
collective_nccl_reducer.cc
|
Ensure that CollectiveParams outlives all references to it.
|
2021-02-04 10:07:20 -08:00 |
|
collective_nccl_reducer.h
|
PR #31485: [ROCm] add ROCm RCCL support
|
2019-08-22 13:08:13 -07:00 |
|
collective_nccl_test.cc
|
Fix collective_nccl_test after breakage in cl/355646163.
|
2021-02-04 13:16:20 -08:00 |
|
collective_nccl.cc
|
Ensure that CollectiveParams outlives all references to it.
|
2021-02-04 10:07:20 -08:00 |
|
collective_nccl.h
|
Make NcclManager part of CollectiveExecutorMgr
|
2020-09-17 14:35:16 -07:00 |
|
collective_ops.cc
|
Remove superfluous ref/unref. We assume AsyncOpKernel outlives its done callback.
|
2021-02-04 11:52:49 -08:00 |
|
compare_and_bitpack_op_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
compare_and_bitpack_op.cc
|
Adding ROCm support for comapre_and_bitpack op
|
2019-05-29 22:13:31 +00:00 |
|
compare_and_bitpack_op.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
concat_lib_cpu.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
concat_lib_cpu.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
concat_lib_gpu_impl.cu.cc
|
Improve GPU dtype coverage of concat
|
2020-09-24 00:52:50 -07:00 |
|
concat_lib_gpu.cc
|
Revert PR #43178: Make fast builds work with MSVC
|
2020-10-05 10:30:22 -07:00 |
|
concat_lib_gpu.h
|
Improve GPU dtype coverage of concat
|
2020-09-24 00:52:50 -07:00 |
|
concat_lib.h
|
Improve GPU dtype coverage of concat
|
2020-09-24 00:52:50 -07:00 |
|
concat_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
concat_op.cc
|
Improve GPU dtype coverage of concat
|
2020-09-24 00:52:50 -07:00 |
|
conditional_accumulator_base_op.cc
|
Boosted Trees: switching to version 2 cond (even when running in v1). This should fix retval issues and enable TF2.0 support
|
2019-06-17 13:41:58 -07:00 |
|
conditional_accumulator_base_op.h
|
|
|
|
conditional_accumulator_base.cc
|
|
|
|
conditional_accumulator_base.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
conditional_accumulator_op.cc
|
Move uses of tensorflow::MakeTypeIndex() to tensorflow::TypeIndex::Make.
|
2020-06-23 12:50:42 -07:00 |
|
conditional_accumulator.h
|
|
|
|
constant_op_gpu.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
constant_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
constant_op.cc
|
Add tf.qint32 support for tf.zeros
|
2021-01-10 05:01:12 +00:00 |
|
constant_op.h
|
Internal change
|
2020-01-10 17:44:33 -08:00 |
|
control_flow_ops_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
control_flow_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
control_flow_ops.h
|
Adds a lowering from Case to _SwitchN+Merge.
|
2019-05-30 18:28:08 -07:00 |
|
conv_2d_gpu_double.cu.cc
|
Add ROCm support for launching 2D GPU convolutions
|
2019-05-30 16:41:42 +00:00 |
|
conv_2d_gpu_float.cu.cc
|
Keep TransformDepth
|
2019-09-06 23:36:33 -07:00 |
|
conv_2d_gpu_half.cu.cc
|
clean-up unused functors
|
2019-09-06 23:36:33 -07:00 |
|
conv_2d_gpu_int_spatial_convolution_backward.cu.cc
|
|
|
|
conv_2d_gpu_int_spatial_convolution.cu.cc
|
Split up conv_2d kernels even further
|
2020-02-06 00:34:58 -08:00 |
|
conv_2d_gpu_int.cu.cc
|
Split up conv_2d kernels even further
|
2020-02-06 00:34:58 -08:00 |
|
conv_2d_gpu_uint8.cu.cc
|
Add ROCm support for launching 2D GPU convolutions
|
2019-05-30 16:41:42 +00:00 |
|
conv_2d_gpu_uint16.cu.cc
|
Add ROCm support for launching 2D GPU convolutions
|
2019-05-30 16:41:42 +00:00 |
|
conv_2d_gpu_uint32.cu.cc
|
Add ROCm support for launching 2D GPU convolutions
|
2019-05-30 16:41:42 +00:00 |
|
conv_2d_gpu_uint64.cu.cc
|
Add ROCm support for launching 2D GPU convolutions
|
2019-05-30 16:41:42 +00:00 |
|
conv_2d_gpu.h
|
PR #44471: [ROCm] Update to use ROCm 3.9 (when building TF with --config=rocm)
|
2020-11-10 00:57:26 -08:00 |
|
conv_2d.h
|
[ROCm] Fix for ROCm CSB Breakage - 200902 - 1
|
2020-09-04 00:41:31 +00:00 |
|
conv_3d.h
|
Use Eigen::CuboidConvolutionBackwardKernel in Conv3DBackpropFilter.
|
2018-09-12 14:21:00 -07:00 |
|
conv_grad_filter_ops_benchmark_test.cc
|
|
|
|
conv_grad_filter_ops.cc
|
Merge pull request #42897 from ROCmSoftwarePlatform:google_upstream_rocm_fix_200902_1
|
2020-09-25 11:54:03 -07:00 |
|
conv_grad_input_ops_benchmark_test.cc
|
|
|
|
conv_grad_input_ops_double.cc
|
Splitting up conv_grad_input_ops.cc to improve build times
|
2020-12-04 05:47:25 -08:00 |
|
conv_grad_input_ops_float.cc
|
Splitting up conv_grad_input_ops.cc to improve build times
|
2020-12-04 05:47:25 -08:00 |
|
conv_grad_input_ops_half.cc
|
Splitting up conv_grad_input_ops.cc to improve build times
|
2020-12-04 05:47:25 -08:00 |
|
conv_grad_input_ops_int32.cc
|
Splitting up conv_grad_input_ops.cc to improve build times
|
2020-12-04 05:47:25 -08:00 |
|
conv_grad_input_ops.cc
|
Splitting up conv_grad_input_ops.cc to improve build times
|
2020-12-04 05:47:25 -08:00 |
|
conv_grad_input_ops.h
|
Replace MKL-DNN v0.x in vanilla TensorFlow and TFRT with DNNL/oneDNN v1.8.
|
2021-02-05 11:35:14 -08:00 |
|
conv_grad_ops_3d.cc
|
Improve convolution error with cudnn 8.0.4.
|
2021-02-02 21:47:55 -08:00 |
|
conv_grad_ops.h
|
Move conv grad shape related utils to a separate lib
|
2019-10-30 13:58:39 -07:00 |
|
conv_grad_shape_utils.cc
|
Improve error message format when reporting batch dimension mismatch in conv2d.
|
2020-07-31 17:26:40 -07:00 |
|
conv_grad_shape_utils.h
|
|
|
|
conv_ops_3d.cc
|
Remove changes made to support TFRT-based OpKernel classes in Conv3d kernel.
|
2020-12-09 16:19:36 -08:00 |
|
conv_ops_benchmark_test.cc
|
Merge pull request #45856 from Intel-tensorflow:mabuzain/fixing-conv-ops-benchmark-test
|
2020-12-21 11:51:46 -08:00 |
|
conv_ops_fused_double.cc
|
Remove EIGEN_USE_LIBXSMM support from Tensorflow.
|
2019-08-19 13:46:32 -07:00 |
|
conv_ops_fused_float.cc
|
Remove EIGEN_USE_LIBXSMM support from Tensorflow.
|
2019-08-19 13:46:32 -07:00 |
|
conv_ops_fused_half.cc
|
Roll-forward:
|
2019-03-07 11:21:33 -08:00 |
|
conv_ops_fused_image_transform.cc
|
fix typos in core directory
|
2020-10-29 02:52:55 +03:00 |
|
conv_ops_fused_impl.h
|
Merge pull request #42897 from ROCmSoftwarePlatform:google_upstream_rocm_fix_200902_1
|
2020-09-25 11:54:03 -07:00 |
|
conv_ops_gpu.h
|
Rollback PR #43237: Integrate cuBLASLt API into backend
|
2020-10-28 17:04:53 -07:00 |
|
conv_ops_test.cc
|
Use TensorFloat-32 instead of TF32 in comments and symbols.
|
2020-09-08 17:38:28 -07:00 |
|
conv_ops_using_gemm.cc
|
Break up tensorflow/core/kernels/BUILD (part 3 of N):
|
2020-08-11 11:04:52 -07:00 |
|
conv_ops.cc
|
Rollback PR #43237: Integrate cuBLASLt API into backend
|
2020-10-28 17:04:53 -07:00 |
|
conv_ops.h
|
Add ROCm support for 2D and 3D convolution ops.
|
2019-06-26 17:03:18 -05:00 |
|
count_ops_test.cc
|
Add tf.sparse.bincount op.
|
2020-04-30 17:50:34 -07:00 |
|
count_ops.cc
|
Fix multiple vulnerabilities in tf.raw_ops.*CountSparseOutput.
|
2020-09-18 18:20:14 -07:00 |
|
count_up_to_op.cc
|
|
|
|
cross_op_gpu.cu.cc
|
Adding ROCm support to cross_op
|
2019-05-30 14:29:45 +00:00 |
|
cross_op_test.cc
|
Remove unused protobuf header inclusions
|
2017-06-06 03:15:27 -07:00 |
|
cross_op.cc
|
Adding ROCm support to cross_op
|
2019-05-30 14:29:45 +00:00 |
|
cross_op.h
|
Fix include guards
|
2018-08-20 11:20:05 -07:00 |
|
ctc_decoder_ops.cc
|
Stops calling std::copy_n over zero-sided vec, which indeed dereferences null.
|
2020-06-15 18:46:51 -07:00 |
|
ctc_loss_op.cc
|
[ROCm] Adding ROCm support for CTC Loss
|
2020-01-27 18:50:24 +00:00 |
|
cudnn_pooling_gpu.cc
|
Adding 3d Pooling using latest MIOpen API
|
2020-06-09 15:11:45 +00:00 |
|
cudnn_pooling_gpu.h
|
Add ROCm support for pooling operators
|
2019-06-28 17:10:51 +00:00 |
|
cudnn_rnn_ops.cc
|
Merge pull request #41832 from lissyx:issue41630
|
2020-08-07 10:08:12 -07:00 |
|
cwise_op_abs.cc
|
[KERNEL_GEN][CPU] Add CPU kernel generation after fixing the build.
|
2021-02-08 07:39:33 -08:00 |
|
cwise_op_acos.cc
|
Enable the generated version of acos and acosh.
|
2021-02-04 08:40:16 -08:00 |
|
cwise_op_acosh.cc
|
Enable the generated version of acos and acosh.
|
2021-02-04 08:40:16 -08:00 |
|
cwise_op_add_1.cc
|
Alias the Add and AddV2 operations in kernel generator generated kernels.
|
2021-02-08 12:24:18 -08:00 |
|
cwise_op_add_2.cc
|
Alias the Add and AddV2 operations in kernel generator generated kernels.
|
2021-02-08 12:24:18 -08:00 |
|
cwise_op_arg.cc
|
Enable the generated kernel for angle.
|
2021-02-04 10:02:22 -08:00 |
|
cwise_op_asin.cc
|
Enable the generated version of asin and asinh.
|
2021-02-04 09:04:21 -08:00 |
|
cwise_op_asinh.cc
|
Enable the generated version of asin and asinh.
|
2021-02-04 09:04:21 -08:00 |
|
cwise_op_atan2.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_atan.cc
|
Enable generated kernels for atanh and atan.
|
2021-01-28 08:53:24 -08:00 |
|
cwise_op_atanh.cc
|
Enable generated kernels for atanh and atan.
|
2021-01-28 08:53:24 -08:00 |
|
cwise_op_bitwise_and.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_bitwise_or.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_bitwise_xor.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_ceil.cc
|
Enable the generated version of ceil.
|
2021-01-29 07:23:01 -08:00 |
|
cwise_op_clip_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
cwise_op_clip.cc
|
Add bfloat16 support to more cwise CPU ops
|
2020-08-06 09:48:01 -07:00 |
|
cwise_op_clip.h
|
Merge changes from github.
|
2018-04-13 17:57:27 -07:00 |
|
cwise_op_complex.cc
|
Enable the generated version of complex and complex_abs.
|
2021-02-01 03:30:50 -08:00 |
|
cwise_op_conj.cc
|
Enable the generated version of conj.
|
2021-02-01 04:24:03 -08:00 |
|
cwise_op_cos.cc
|
Enable the generated version of cos and cosh kernels.
|
2021-01-29 03:47:34 -08:00 |
|
cwise_op_cosh.cc
|
Enable the generated version of cos and cosh kernels.
|
2021-01-29 03:47:34 -08:00 |
|
cwise_op_digamma.cc
|
Enable generated versions of lgamma and digamma.
|
2021-02-08 11:30:33 -08:00 |
|
cwise_op_div.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_equal_to_1.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_equal_to_2.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_erf.cc
|
Enable the generated version of erf and erfc.
|
2021-02-01 07:23:37 -08:00 |
|
cwise_op_erfc.cc
|
Enable the generated version of erf and erfc.
|
2021-02-01 07:23:37 -08:00 |
|
cwise_op_exp.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_expm1.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_floor_div.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_floor_mod.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_floor.cc
|
Enable the generated version of floor.
|
2021-02-01 08:15:59 -08:00 |
|
cwise_op_gpu_abs.cu.cc
|
Enable the generated version of complex and complex_abs.
|
2021-02-01 03:30:50 -08:00 |
|
cwise_op_gpu_acos.cu.cc
|
Enable the generated version of acos and acosh.
|
2021-02-04 08:40:16 -08:00 |
|
cwise_op_gpu_acosh.cu.cc
|
Enable the generated version of acos and acosh.
|
2021-02-04 08:40:16 -08:00 |
|
cwise_op_gpu_add.cu.cc
|
Alias the Add and AddV2 operations in kernel generator generated kernels.
|
2021-02-08 12:24:18 -08:00 |
|
cwise_op_gpu_arg.cu.cc
|
Enable the generated kernel for angle.
|
2021-02-04 10:02:22 -08:00 |
|
cwise_op_gpu_asin.cu.cc
|
Enable the generated version of asin and asinh.
|
2021-02-04 09:04:21 -08:00 |
|
cwise_op_gpu_asinh.cu.cc
|
Enable the generated version of asin and asinh.
|
2021-02-04 09:04:21 -08:00 |
|
cwise_op_gpu_atan2.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_atan.cu.cc
|
Enable generated kernels for atanh and atan.
|
2021-01-28 08:53:24 -08:00 |
|
cwise_op_gpu_atanh.cu.cc
|
Enable generated kernels for atanh and atan.
|
2021-01-28 08:53:24 -08:00 |
|
cwise_op_gpu_bitwise_and.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_bitwise_or.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_bitwise_xor.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_ceil.cu.cc
|
Enable the generated version of ceil.
|
2021-01-29 07:23:01 -08:00 |
|
cwise_op_gpu_complex.cu.cc
|
Enable the generated version of complex and complex_abs.
|
2021-02-01 03:30:50 -08:00 |
|
cwise_op_gpu_conj.cu.cc
|
Enable the generated version of conj.
|
2021-02-01 04:24:03 -08:00 |
|
cwise_op_gpu_cos.cu.cc
|
Enable the generated version of cos and cosh kernels.
|
2021-01-29 03:47:34 -08:00 |
|
cwise_op_gpu_cosh.cu.cc
|
Enable the generated version of cos and cosh kernels.
|
2021-01-29 03:47:34 -08:00 |
|
cwise_op_gpu_digamma.cu.cc
|
Enable generated versions of lgamma and digamma.
|
2021-02-08 11:30:33 -08:00 |
|
cwise_op_gpu_div.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_equal_to.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_erf.cu.cc
|
Enable the generated version of erf and erfc.
|
2021-02-01 07:23:37 -08:00 |
|
cwise_op_gpu_erfc.cu.cc
|
Enable the generated version of erf and erfc.
|
2021-02-01 07:23:37 -08:00 |
|
cwise_op_gpu_exp.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_expm1.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_floor_div.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_floor_mod.cu.cc
|
|
|
|
cwise_op_gpu_floor.cu.cc
|
Enable the generated version of floor.
|
2021-02-01 08:15:59 -08:00 |
|
cwise_op_gpu_greater_equal.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_greater.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_igammas_double.cu.cc
|
|
|
|
cwise_op_gpu_igammas_float.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_imag.cu.cc
|
Enable the generated version of real and imag kernels.
|
2021-02-01 09:52:13 -08:00 |
|
cwise_op_gpu_inverse.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_invert.cu.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_gpu_isfinite.cu.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_gpu_isinf.cu.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_gpu_isnan.cu.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_gpu_left_shift.cu.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_gpu_less_equal.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_less.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_lgamma.cu.cc
|
Enable generated versions of lgamma and digamma.
|
2021-02-08 11:30:33 -08:00 |
|
cwise_op_gpu_log1p.cu.cc
|
Enable the generated kernel for log and log1p.
|
2021-02-01 13:41:33 -08:00 |
|
cwise_op_gpu_log.cu.cc
|
Enable the generated kernel for log and log1p.
|
2021-02-01 13:41:33 -08:00 |
|
cwise_op_gpu_logical_and.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_logical_not.cu.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_gpu_logical_or.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_maximum.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_minimum.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_mod.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_mul.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_neg.cu.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_gpu_not_equal_to.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_pow.cu.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_gpu_random_grad.cu.cc
|
|
|
|
cwise_op_gpu_real.cu.cc
|
Enable the generated version of real and imag kernels.
|
2021-02-01 09:52:13 -08:00 |
|
cwise_op_gpu_right_shift.cu.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_gpu_rint.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_round.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_rsqrt.cu.cc
|
Enable the generated version of rsqrt.
|
2021-02-01 14:01:26 -08:00 |
|
cwise_op_gpu_select.cu.cc
|
Let SelectV2 handle dimensions up to 8. This matches recent changes to Where.
|
2019-06-26 15:34:43 -07:00 |
|
cwise_op_gpu_sigmoid.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_sign.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_sin.cu.cc
|
Enable the generated version of sin and sinh.
|
2021-02-04 09:41:22 -08:00 |
|
cwise_op_gpu_sinh.cu.cc
|
Enable the generated version of sin and sinh.
|
2021-02-04 09:41:22 -08:00 |
|
cwise_op_gpu_sqrt.cu.cc
|
Enable the generated versions of square and sqrt.
|
2021-01-29 04:23:04 -08:00 |
|
cwise_op_gpu_square.cu.cc
|
Enable the generated versions of square and sqrt.
|
2021-01-29 04:23:04 -08:00 |
|
cwise_op_gpu_squared_difference.cu.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_gpu_sub.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_gpu_tan.cu.cc
|
Enable the generated version of tan kernel.
|
2021-01-28 10:23:21 -08:00 |
|
cwise_op_gpu_tanh.cu.cc
|
Add support for f16 generated tanh kernel.
|
2020-06-15 10:12:37 -07:00 |
|
cwise_op_gpu_xdivy.cu.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_gpu_xlog1py.cu.cc
|
Add tf.math.xlog1py, a safe way to compute x * log1p(y)
|
2020-01-09 14:28:37 -08:00 |
|
cwise_op_gpu_xlogy.cu.cc
|
Update cwise_op_gpu_xlogy.cu.cc
|
2019-08-15 10:49:49 -05:00 |
|
cwise_op_gpu_zeta.cu.cc
|
Add appropriate preprocessor defines to avoid building Eigen kernels when building with kernel generator enabled.
|
2021-02-09 03:29:44 -08:00 |
|
cwise_op_greater_equal.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_greater.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_igammas.cc
|
Add ROCm support for most cwise ops
|
2019-05-31 19:13:19 +00:00 |
|
cwise_op_imag.cc
|
Enable the generated version of real and imag kernels.
|
2021-02-01 09:52:13 -08:00 |
|
cwise_op_invert.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_isfinite.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_isinf.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_isnan.cc
|
Enable generated versions of is_inf, is_nan and is_finite.
|
2021-02-02 01:18:07 -08:00 |
|
cwise_op_left_shift.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_less_equal.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_less.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_lgamma.cc
|
Enable generated versions of lgamma and digamma.
|
2021-02-08 11:30:33 -08:00 |
|
cwise_op_log1p.cc
|
Enable the generated kernel for log and log1p.
|
2021-02-01 13:41:33 -08:00 |
|
cwise_op_log.cc
|
Enable the generated kernel for log and log1p.
|
2021-02-01 13:41:33 -08:00 |
|
cwise_op_logical_and.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_logical_not.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_logical_or.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_maximum.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_minimum.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_mod.cc
|
|
|
|
cwise_op_mul_1.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_mul_2.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_ndtri_double.cu.cc
|
Remove ErfinvGrad and NdtriGrad since gradients are already registerd in math_grad.cc
|
2019-10-03 10:47:01 -07:00 |
|
cwise_op_ndtri_float.cu.cc
|
Remove ErfinvGrad and NdtriGrad since gradients are already registerd in math_grad.cc
|
2019-10-03 10:47:01 -07:00 |
|
cwise_op_ndtri.cc
|
Remove ErfinvGrad and NdtriGrad since gradients are already registerd in math_grad.cc
|
2019-10-03 10:47:01 -07:00 |
|
cwise_op_neg_1.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_neg_2.cc
|
Enable the generated versions of invert, logical_not and neg.
|
2021-02-01 12:19:33 -08:00 |
|
cwise_op_not_equal_to_1.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_not_equal_to_2.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_pow.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_random_grad.cc
|
|
|
|
cwise_op_real.cc
|
Enable the generated version of real and imag kernels.
|
2021-02-01 09:52:13 -08:00 |
|
cwise_op_reciprocal.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_right_shift.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_rint.cc
|
Add int32 support to floor, ceil & rint
|
2019-12-04 18:47:55 -08:00 |
|
cwise_op_round.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_rsqrt.cc
|
Enable the generated version of rsqrt.
|
2021-02-01 14:01:26 -08:00 |
|
cwise_op_select.cc
|
Cleanup Selectv2 broadcasting
|
2020-10-09 10:13:47 +01:00 |
|
cwise_op_sigmoid.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_sign.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_sin.cc
|
Enable the generated version of sin and sinh.
|
2021-02-04 09:41:22 -08:00 |
|
cwise_op_sinh.cc
|
Enable the generated version of sin and sinh.
|
2021-02-04 09:41:22 -08:00 |
|
cwise_op_sqrt.cc
|
Enable the generated versions of square and sqrt.
|
2021-01-29 04:23:04 -08:00 |
|
cwise_op_square.cc
|
Enable the generated versions of square and sqrt.
|
2021-01-29 04:23:04 -08:00 |
|
cwise_op_squared_difference.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_sub.cc
|
[KERNEL_GEN] Rename MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED -> MLIR_GENERATED_EXPERIMENTAL_KERNELS_ENABLED.
|
2021-02-03 04:47:35 -08:00 |
|
cwise_op_tan.cc
|
Enable the generated version of tan kernel.
|
2021-01-28 10:23:21 -08:00 |
|
cwise_op_tanh.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_xdivy.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_xlog1py.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_xlogy.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_op_zeta.cc
|
Add kernel definition for zeta operation.
|
2021-02-04 01:30:42 -08:00 |
|
cwise_ops_common.cc
|
Add incompatible_shape_error attribute to equal op
|
2019-09-05 16:00:40 -07:00 |
|
cwise_ops_common.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_ops_gpu_common.cu.h
|
Merge pull request #42689 from ROCmSoftwarePlatform:google_upstream_switch_to_rocm37
|
2020-09-25 23:37:01 -07:00 |
|
cwise_ops_gpu_gradients.cu.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
cwise_ops_gradients.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
cwise_ops_test.cc
|
Fix and re-enable cwise ops benchmark.
|
2020-10-30 16:12:39 -07:00 |
|
cwise_ops.h
|
[mlir] Lower tf.Rint to tf.Round, as they're the same.
|
2021-01-26 01:00:29 -08:00 |
|
data_format_ops_gpu.cu.cc
|
Adding ROCm support for data format ops
|
2019-05-28 20:37:44 +00:00 |
|
data_format_ops.cc
|
Validate that DataFormat* attributes form a permutation.
|
2020-12-07 11:25:52 -08:00 |
|
data_format_ops.h
|
Merge pull request #43226 from kaixih:reduce_ops_layout
|
2020-09-25 12:56:20 -07:00 |
|
debug_ops_gpu.cu.cc
|
[tfdbg] Fix DebugNumericSummaryV2OP kernel mode=CURT_HEALTH breakage on Windows GPU builds
|
2019-12-20 08:53:47 -08:00 |
|
debug_ops_test.cc
|
[tfdbg] Remove use of dirent.h in debug_ops_test.cc
|
2020-02-02 19:14:32 -08:00 |
|
debug_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
debug_ops.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
decode_compressed_op.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
decode_csv_op.cc
|
|
|
|
decode_padded_raw_op.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
decode_proto_op.cc
|
Fixes/updates for tensorflow::tstring migration.
|
2019-11-26 23:35:38 -08:00 |
|
decode_raw_op.cc
|
[tstring] Avoid temporary std::string creation in DecodeRawOp.
|
2020-01-23 14:48:29 -08:00 |
|
decode_wav_op_test.cc
|
Migrated a subset of kernels to use tstring.
|
2019-08-14 09:29:06 -07:00 |
|
decode_wav_op.cc
|
Fix some unnecessary string copies when reading from or writing to a Tensor.
|
2019-08-08 15:38:48 -07:00 |
|
deep_conv2d_test.cc
|
Experimental Conv2D implementation specialized for deep convolutions (i.e. large in_depth * out_depth, see cost model in deep_conv2d.cc for details).
|
2016-08-31 11:21:41 -07:00 |
|
deep_conv2d.cc
|
Change some const declarations to constexpr
|
2020-04-20 14:59:17 -07:00 |
|
deep_conv2d.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
dense_update_functor_gpu.cu.cc
|
Revert PR #43178: Make fast builds work with MSVC
|
2020-10-05 10:30:22 -07:00 |
|
dense_update_functor.cc
|
Automated rollback of commit 4485ff2c9f
|
2019-08-12 15:33:25 -07:00 |
|
dense_update_functor.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
dense_update_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
depthtospace_op_gpu.cu.cc
|
Revert PR #43178: Make fast builds work with MSVC
|
2020-10-05 10:30:22 -07:00 |
|
depthtospace_op.cc
|
rephrase wording
|
2020-08-31 16:42:37 +02:00 |
|
depthtospace_op.h
|
Implement NCHW_VECT_C support for tf.depth_to_space on GPU.
|
2017-10-11 18:20:06 -07:00 |
|
depthwise_conv_grad_op.cc
|
Remove CanUseCudnn and CudnnConvComputeMode functions.
|
2020-08-18 12:04:59 -07:00 |
|
depthwise_conv_op_gpu_double.cu.cc
|
[ROCm] Adding support to depthwise_conv_op
|
2019-07-09 12:19:23 -05:00 |
|
depthwise_conv_op_gpu_float.cu.cc
|
[ROCm] Adding support to depthwise_conv_op
|
2019-07-09 12:19:23 -05:00 |
|
depthwise_conv_op_gpu_half.cu.cc
|
|
|
|
depthwise_conv_op_gpu.h
|
fix LaunchDepthwiseConvBackpropFilterOp
|
2020-06-16 23:38:53 +08:00 |
|
depthwise_conv_op.cc
|
Remove CanUseCudnn and CudnnConvComputeMode functions.
|
2020-08-18 12:04:59 -07:00 |
|
depthwise_conv_op.h
|
Optimize depthwise convolution. Pscatter is somewhat slow,
|
2020-12-08 12:21:42 -08:00 |
|
depthwise_conv_ops_test.cc
|
[ROCm] Re-enabling unit-tests that are now passing on ROCm platform
|
2020-12-07 16:41:10 +00:00 |
|
dequantize_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
dequantize_op.cc
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
deserialize_sparse_string_op.cc
|
|
|
|
deserialize_sparse_variant_op.cc
|
|
|
|
diag_op_gpu.cu.cc
|
|
|
|
diag_op_test.cc
|
Internal tests cleanup.
|
2020-10-27 21:28:44 -07:00 |
|
diag_op.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
diag_op.h
|
|
|
|
dilation_ops_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
dilation_ops.cc
|
Splitting out common_shape_fns.cc:
|
2020-03-03 14:28:49 -08:00 |
|
dilation_ops.h
|
Morphological filtering operations: Dilation and erosion.
|
2016-06-02 20:48:14 -07:00 |
|
dynamic_partition_op_gpu.cu.cc
|
Support ROCm in dynamic_parition fix
|
2020-08-24 00:26:22 -07:00 |
|
dynamic_partition_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
dynamic_partition_op.cc
|
Rollforward: Add uint32 & uint64 to TF_CALL_INTEGRAL_TYPES
|
2020-06-22 23:09:59 -07:00 |
|
dynamic_stitch_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
dynamic_stitch_op_test.cc
|
Use absl instead of deprecated str_util
|
2019-05-21 08:10:10 -07:00 |
|
dynamic_stitch_op.cc
|
Change the type of slice_size from int to auto to avoid overflow.
|
2020-09-25 13:19:34 -07:00 |
|
edit_distance_op.cc
|
|
|
|
eigen_activations_test.cc
|
Add benchmark for the SpatialConvolution specialized gemm_pack_rhs.
|
2018-11-02 12:14:43 -07:00 |
|
eigen_activations.h
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
eigen_attention_test.cc
|
Add benchmark for the SpatialConvolution specialized gemm_pack_rhs.
|
2018-11-02 12:14:43 -07:00 |
|
eigen_attention.h
|
Merge pull request #38549 from yongtang:12829-extract_glimpse
|
2020-05-21 12:14:19 -07:00 |
|
eigen_backward_cuboid_convolutions_test.cc
|
|
|
|
eigen_backward_cuboid_convolutions.h
|
Use Eigen::CuboidConvolutionBackwardKernel in Conv3DBackpropFilter.
|
2018-09-12 14:21:00 -07:00 |
|
eigen_backward_spatial_convolutions_test.cc
|
Split eigen_backward_spatial_convolutions_test into two.
|
2020-01-24 16:07:33 -08:00 |
|
eigen_backward_spatial_convolutions.h
|
|
|
|
eigen_benchmark_cpu_test.cc
|
|
|
|
eigen_benchmark.h
|
|
|
|
eigen_contraction_kernel.cc
|
|
|
|
eigen_contraction_kernel.h
|
Replace MKL-DNN v0.x in vanilla TensorFlow and TFRT with DNNL/oneDNN v1.8.
|
2021-02-05 11:35:14 -08:00 |
|
eigen_convolution_helpers.h
|
Change some const declarations to constexpr
|
2020-04-20 14:59:17 -07:00 |
|
eigen_cuboid_convolution.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
eigen_mkldnn_contraction_kernel_test.cc
|
Replace MKL-DNN v0.x in vanilla TensorFlow and TFRT with DNNL/oneDNN v1.8.
|
2021-02-05 11:35:14 -08:00 |
|
eigen_pooling_test.cc
|
|
|
|
eigen_pooling.h
|
Update Eigen to
|
2020-10-02 20:28:38 -07:00 |
|
eigen_spatial_convolutions_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
eigen_spatial_convolutions-inl.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
eigen_spatial_convolutions.h
|
Default initialize offsets to avoid UB
|
2020-12-21 12:48:29 -08:00 |
|
eigen_volume_patch.h
|
Change some const declarations to constexpr
|
2020-04-20 14:59:17 -07:00 |
|
encode_proto_op.cc
|
Update kernels and related libs to use tstring.
|
2019-08-27 18:22:08 -07:00 |
|
encode_wav_op_test.cc
|
Remove using directives. Test appertaining to ops have been moved into namespace tensorflow::ops; all other tests now use explicit using-declarations.
|
2018-01-02 10:49:13 -08:00 |
|
encode_wav_op.cc
|
Update kernels and related libs to use tstring.
|
2019-08-27 18:22:08 -07:00 |
|
example_parsing_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
example_parsing_ops.cc
|
Fix a bug where vector::reserve was being called on config.sparse instead of config.ragged.
|
2020-10-14 11:59:47 -07:00 |
|
fact_op.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
fake_quant_ops_functor.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
fake_quant_ops_gpu.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
fake_quant_ops_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
fake_quant_ops.cc
|
PR #36526: Removed NDIMS template arg from BinaryElementWiseOp::Operate
|
2020-02-26 20:38:14 -08:00 |
|
fft_ops.cc
|
Rollback PR #43237: Integrate cuBLASLt API into backend
|
2020-10-28 17:04:53 -07:00 |
|
fifo_queue_op.cc
|
|
|
|
fifo_queue.cc
|
|
|
|
fifo_queue.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
fill_functor.cc
|
Add tf.qint32 support for tf.zeros
|
2021-01-10 05:01:12 +00:00 |
|
fill_functor.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
fill_functor.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
fingerprint_op_test.cc
|
Fixing a crash in tf.fingerprint([]). We now just return []
|
2020-09-30 08:09:56 -07:00 |
|
fingerprint_op.cc
|
Fixing a crash in tf.fingerprint([]). We now just return []
|
2020-09-30 08:09:56 -07:00 |
|
fixed_length_record_reader_op.cc
|
|
|
|
fractional_avg_pool_op.cc
|
|
|
|
fractional_max_pool_op.cc
|
|
|
|
fractional_pool_common.cc
|
|
|
|
fractional_pool_common.h
|
|
|
|
function_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
function_ops.h
|
Use TraceMeOp and TraceMeEncode to hide TraceString encoding details
|
2020-07-24 10:59:15 -07:00 |
|
functional_ops.cc
|
Replace deprecated thread annotations macros.
|
2021-01-25 11:07:15 -08:00 |
|
fused_batch_norm_ex_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
fused_batch_norm_op_test.cc
|
Automated rollback of commit 80acd88cd4
|
2020-02-18 13:51:51 -08:00 |
|
fused_batch_norm_op.cc
|
replace alternative operator to fix msvc build break
|
2021-01-24 23:01:48 +08:00 |
|
fused_batch_norm_op.cu.cc
|
Merged commit includes the following changes:
|
2020-07-02 10:46:26 -07:00 |
|
fused_batch_norm_op.h
|
Qualify uses of std::string
|
2020-07-31 13:57:36 -07:00 |
|
fused_eigen_output_kernels.cc
|
Enable conv + (bias/bn) + leakyrelu fusion
|
2020-08-21 15:47:17 +08:00 |
|
fused_eigen_output_kernels.h
|
Merge pull request #42551 from Intel-tensorflow:yunfeimao/leaky_relu_fusion_eigen2
|
2020-08-28 08:16:17 -07:00 |
|
gather_functor_batched_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_functor_batched_gpu.cu.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
gather_functor_batched.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_functor_batched.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
gather_functor_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_functor_gpu.cu.h
|
adjust gather ops launch config. for NCF model, this means ~20% gain. (due to grid size from 80->160 on volta).
|
2020-05-19 16:12:30 -07:00 |
|
gather_functor.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_functor.h
|
Fix HandleCopies so it no longer invokes UB when the params or out TensorMaps are empty with no backing data.
|
2020-05-28 17:01:07 -07:00 |
|
gather_nd_op_cpu_impl_0.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_1.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_2.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_3.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_4.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_5.cc
|
Split gather_nd CPU functors into multiple files for faster compile times.
|
2016-08-02 16:17:47 -07:00 |
|
gather_nd_op_cpu_impl_6.cc
|
Make gather_nd & scatter_nd handle rank 6 and 7 indices.
|
2017-12-04 01:00:36 -08:00 |
|
gather_nd_op_cpu_impl_7.cc
|
Make gather_nd & scatter_nd handle rank 6 and 7 indices.
|
2017-12-04 01:00:36 -08:00 |
|
gather_nd_op_cpu_impl.h
|
Use parallelFor in GatherNdSlice. Saw speedups from 3-12% on Intel Skylake.
|
2019-07-11 17:23:44 -07:00 |
|
gather_nd_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_nd_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
gather_nd_op.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
gather_nd_op.h
|
Print node name in GatherNdSlice error msg.
|
2020-12-10 00:32:00 -08:00 |
|
gather_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
gather_op.cc
|
Fix data race in GatherOp::Compute().
|
2020-12-15 12:17:14 -08:00 |
|
gemm_functors.h
|
Use mkldnn sgemm in Eigen contractions.
|
2018-11-09 15:24:28 -08:00 |
|
generate_vocab_remapping_op.cc
|
|
|
|
gpu_device_array_gpu.h
|
[ROCm] Misc updates for the ROCm platform
|
2019-07-11 20:38:46 +00:00 |
|
gpu_device_array.h
|
Remove unused unique_tensor_references library.
|
2020-03-12 16:35:29 -07:00 |
|
gpu_prim.h
|
Merge pull request #39429 from ekuznetsov139:google-upstream-topk
|
2020-10-18 22:12:36 -07:00 |
|
gpu_utils.cc
|
Rollback PR #43237: Integrate cuBLASLt API into backend
|
2020-10-28 17:04:53 -07:00 |
|
gpu_utils.h
|
Log value of TF_AUTOTUNE_THRESHOLD
|
2020-11-06 12:33:42 -08:00 |
|
guarantee_const_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
guarantee_const_op.cc
|
Adds a GuaranteeConstOp.
|
2017-12-01 16:29:19 -08:00 |
|
hinge-loss.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
histogram_op_gpu.cu.cc
|
Merge pull request #36640 from ROCmSoftwarePlatform:google-upstream-gpuprim
|
2020-04-15 09:18:52 -07:00 |
|
histogram_op.cc
|
Segmentation Fault 11 thrown by tensorflow::HistogramFixedWidthFunctor::Compute in histogram_op.cc
|
2019-07-11 11:47:41 -07:00 |
|
histogram_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
host_constant_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
host_constant_op.h
|
Internal change
|
2020-01-10 17:44:33 -08:00 |
|
i_remote_fused_graph_executor.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
i_remote_fused_graph_ops_definitions.cc
|
Add a way to fuse a graph by remote graph executor so that users don't need to be aware of supported op types, node names, subgraph stracture etc.
|
2017-07-18 15:23:57 -07:00 |
|
i_remote_fused_graph_ops_definitions.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
identity_n_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
identity_n_op.cc
|
Add required TPU_SYSTEM device kernel registrations for TPUs
|
2020-07-01 11:48:41 -07:00 |
|
identity_n_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
identity_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
identity_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
identity_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
identity_reader_op.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
immutable_constant_op_test.cc
|
Add TF_USE_FILESYSTEM_METHODS_WITH_NO_TRANSACTION_SUPPORT macro to modified classes
|
2020-08-04 23:20:15 -07:00 |
|
immutable_constant_op.cc
|
Allowlist certain data types to avoid a seg fault.
|
2021-02-08 12:37:40 -08:00 |
|
immutable_constant_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
in_topk_op_gpu.cu.cc
|
|
|
|
in_topk_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
in_topk_op.cc
|
[ROCm] enable InTopK op on ROCm.
|
2019-08-13 16:31:25 +00:00 |
|
in_topk_op.h
|
[ROCm] enable InTopK op on ROCm.
|
2019-08-13 16:31:25 +00:00 |
|
initializable_lookup_table.cc
|
[-Wsign-compare] warning fixes batch 4
|
2020-06-24 21:58:00 +00:00 |
|
initializable_lookup_table.h
|
|
|
|
inplace_ops_functor_gpu.cu.cc
|
Add complex64 and complex128 gpu support for tensor_scatter_nd_add
|
2020-06-18 16:17:52 +00:00 |
|
inplace_ops_functor.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
inplace_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
isotonic_regression_op_test.cc
|
Fix kokoro build, add guard in IsotonicreRegressionOp benchmark
|
2020-08-18 13:30:13 -07:00 |
|
isotonic_regression_op.cc
|
Differentiable isotonic regression in TensorFlow.
|
2020-08-18 10:48:08 -07:00 |
|
kernel_platform_strings.h
|
|
|
|
l2loss_op_gpu.cu.cc
|
Merge pull request #36640 from ROCmSoftwarePlatform:google-upstream-gpuprim
|
2020-04-15 09:18:52 -07:00 |
|
l2loss_op.cc
|
|
|
|
l2loss_op.h
|
|
|
|
linalg_ops_common.h
|
Break up core/kernels/BUILD (part 1 of N):
|
2020-08-04 17:24:28 -07:00 |
|
list_kernels.cc
|
Define TensorList class in a separate library
|
2020-01-24 17:37:07 -08:00 |
|
list_kernels.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
list_kernels.h
|
Make element_shape_except_first_dim as a local variable instead of instance variable.
|
2020-11-24 10:12:32 -08:00 |
|
listdiff_op.cc
|
|
|
|
lmdb_reader_op.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
load_and_remap_matrix_op.cc
|
|
|
|
logging_ops_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
logging_ops.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
logging_ops.h
|
Add support for AssertOp in XLA
|
2019-05-20 13:30:08 -07:00 |
|
logistic-loss.h
|
|
|
|
lookup_table_init_op.cc
|
[tf.data] Enforcing usage of the tensorflow::data namespace for tf.data C++ classes.
|
2020-08-24 08:10:41 -07:00 |
|
lookup_table_init_op.h
|
|
|
|
lookup_table_op.cc
|
Merge pull request #43269 from rhdong:RFC237-patch-for-recsys-sig
|
2020-10-27 08:06:45 -07:00 |
|
lookup_table_op.h
|
|
|
|
lookup_util.cc
|
[tf.data] Enforcing usage of the tensorflow::data namespace for tf.data C++ classes.
|
2020-08-24 08:10:41 -07:00 |
|
lookup_util.h
|
Modify InitializeTableFromDatasetOp to be async and add the table init op to the table initializers collection
|
2020-07-01 02:53:55 -07:00 |
|
loss_test.cc
|
|
|
|
loss.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
lrn_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
lrn_op.cc
|
Adding ROCm support for the LRN op
|
2019-07-26 21:15:11 +00:00 |
|
map_kernels.cc
|
rename ListKeys to StackKeys
|
2020-08-13 18:57:03 +00:00 |
|
map_kernels.h
|
Merge pull request #42300 from kttian:list_keys
|
2020-08-25 10:12:32 -07:00 |
|
map_stage_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
matching_files_op.cc
|
Fix asan error, when there are no matching files in matching_files_op.
|
2020-08-03 14:21:09 -07:00 |
|
matmul_op_complex.cc
|
Clean up duplicated code in matmul_op/batch_matmul_op and associated tests. Having a separate redundant implementation for non-batch matmul is a maintenance burden.
|
2020-10-29 11:58:12 -07:00 |
|
matmul_op_fused.cc
|
Fix code format
|
2020-12-23 16:47:36 +08:00 |
|
matmul_op_impl.h
|
Fix regression in (non-batch) matrix multiplication:
|
2020-11-17 15:58:05 -08:00 |
|
matmul_op_real.cc
|
Clean up duplicated code in matmul_op/batch_matmul_op and associated tests. Having a separate redundant implementation for non-batch matmul is a maintenance burden.
|
2020-10-29 11:58:12 -07:00 |
|
matmul_op_test.cc
|
Enable Eigen MatMul + Bias + LeakyRelu fusion
|
2020-12-21 10:47:02 +08:00 |
|
matmul_op.h
|
Adding ROCm support for the "matmul" op
|
2019-05-13 18:33:22 +00:00 |
|
maxpooling_op_gpu.cu.cc
|
fix typos in core directory
|
2020-10-29 02:52:55 +03:00 |
|
maxpooling_op_gpu.h
|
Remove CanUseCudnn and CudnnConvComputeMode functions.
|
2020-08-18 12:04:59 -07:00 |
|
maxpooling_op.cc
|
Support int32 argmax output in MaxPoolWithArgmax
|
2020-12-15 19:00:07 -08:00 |
|
maxpooling_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
merge_v2_checkpoints_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
meta_support.cc
|
Make DebugString() in ResourceBase constant
|
2019-01-08 02:28:38 -08:00 |
|
meta_support.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
mfcc_dct_test.cc
|
Remove some unused C++ variables in core/.
|
2019-02-25 17:04:39 -08:00 |
|
mfcc_dct.cc
|
Add Mfcc op to TensorFlow for speech feature generation
|
2017-04-21 11:24:49 -07:00 |
|
mfcc_dct.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
mfcc_mel_filterbank_test.cc
|
|
|
|
mfcc_mel_filterbank.cc
|
|
|
|
mfcc_mel_filterbank.h
|
|
|
|
mfcc_op_test.cc
|
|
|
|
mfcc_op.cc
|
|
|
|
mfcc_test.cc
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
mfcc.cc
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
mfcc.h
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
multinomial_op_gpu.cu.cc
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
multinomial_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
multinomial_op.cc
|
Adding ROCm support for the multinomial op
|
2019-06-25 15:08:11 +00:00 |
|
multinomial_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
mutex_ops.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
nccl_ops.cc
|
Merge pull request #31483 from ROCmSoftwarePlatform:google_upstream_nccl_stream_early
|
2019-09-05 11:08:06 -07:00 |
|
nextafter_op_gpu.cu.cc
|
[ROCm] enable nextafter op on ROCm.
|
2019-08-15 10:37:02 -05:00 |
|
nextafter_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
nextafter_op.h
|
[ROCm] enable nextafter op on ROCm.
|
2019-08-15 10:37:02 -05:00 |
|
nn_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
no_op.cc
|
Add required TPU_SYSTEM device kernel registrations for TPUs
|
2020-07-01 11:48:41 -07:00 |
|
no_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
nth_element_op.cc
|
Prevent integer truncation from 64 to 32 bits.
|
2020-09-18 17:56:30 -07:00 |
|
nth_element_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
one_hot_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
one_hot_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
one_hot_op.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
one_hot_op.h
|
Optimize OneHot op on CPU
|
2019-10-22 12:49:09 -07:00 |
|
ops_testutil_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
ops_testutil.cc
|
Fixing and enabling NonMaxSuppression for ROCm
|
2020-06-26 19:32:26 +00:00 |
|
ops_testutil.h
|
Move uses of tensorflow::MakeTypeIndex() to tensorflow::TypeIndex::Make.
|
2020-06-23 12:50:42 -07:00 |
|
ops_util_test.cc
|
Splitting out common_shape_fns.cc:
|
2020-03-03 14:28:49 -08:00 |
|
ops_util.h
|
Move ops_util to core/framework.
|
2018-10-22 10:20:57 -07:00 |
|
pack_op.cc
|
Internal change
|
2021-01-19 15:03:26 -08:00 |
|
pad_op_gpu.cu.cc
|
Increase the supported number of dimensions for slice, strided_slice and pad to 8
|
2019-08-22 16:54:18 -07:00 |
|
pad_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
pad_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
padding_fifo_queue_op.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
padding_fifo_queue.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
padding_fifo_queue.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
parameterized_truncated_normal_op_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
parameterized_truncated_normal_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
parameterized_truncated_normal_op.cc
|
Prevent integer truncation from 64 to 32 bits.
|
2020-09-18 17:56:30 -07:00 |
|
parameterized_truncated_normal_op.h
|
Add StatelessParameterizedTruncatedNormal sampler.
|
2020-06-18 10:28:38 -07:00 |
|
parse_tensor_op.cc
|
Migrated a subset of kernels to use tstring.
|
2019-08-14 09:29:06 -07:00 |
|
parse_tensor_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
partitioned_function_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
partitioned_function_ops.h
|
Add ability for functions to share rendezvous
|
2020-06-08 11:47:46 -07:00 |
|
poisson-loss.h
|
Add the poisson log loss to the SDCA optimizer.
|
2018-08-31 11:35:21 -07:00 |
|
pooling_ops_3d_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
pooling_ops_3d_gpu.h
|
Add ROCm support for pooling operators
|
2019-06-28 17:10:51 +00:00 |
|
pooling_ops_3d.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
pooling_ops_3d.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
pooling_ops_common_gpu.h
|
Rollforward of 93c22a269b. Windows and RCOM builds have been fixed.
|
2020-08-31 14:33:25 -07:00 |
|
pooling_ops_common.cc
|
[ROCm] Fix for ROCm CSB Breakage - 200902 - 2
|
2020-09-28 13:32:21 +00:00 |
|
pooling_ops_common.h
|
Rollforward of 93c22a269b. Windows and RCOM builds have been fixed.
|
2020-08-31 14:33:25 -07:00 |
|
population_count_op_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
population_count_op.cc
|
|
|
|
population_count_op.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
priority_queue_op.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
priority_queue.cc
|
|
|
|
priority_queue.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
quantization_utils_test.cc
|
Internal change
|
2020-03-05 11:36:56 -08:00 |
|
quantization_utils.cc
|
Move contrib/quantization ops to tensorflow/core
|
2016-10-17 16:34:02 -07:00 |
|
quantization_utils.h
|
Update Eigen to
|
2020-10-02 20:28:38 -07:00 |
|
quantize_and_dequantize_op_gpu.cu.cc
|
Create a V2 Op to stop the gradient when the input is out of range.
|
2020-10-12 10:39:36 -07:00 |
|
quantize_and_dequantize_op_test.cc
|
Merge pull request #44408 from Molkree:typos_in_core_dir
|
2020-11-06 15:01:58 -08:00 |
|
quantize_and_dequantize_op.cc
|
Prevent segfault in quantize_and_dequantize
|
2020-10-19 18:03:39 -07:00 |
|
quantize_and_dequantize_op.h
|
Create a V2 Op to stop the gradient when the input is out of range.
|
2020-10-12 10:39:36 -07:00 |
|
quantize_down_and_shrink_range_op_test.cc
|
|
|
|
quantize_down_and_shrink_range.cc
|
|
|
|
quantize_op_test.cc
|
Remove uses of rand_r in quantize/dequantize op_test.
|
2020-01-11 14:35:01 -08:00 |
|
quantize_op.cc
|
Rename scalar_round_op_google -> scalar_round_half_to_even_op to document its semantics.
|
2019-12-17 12:47:10 -08:00 |
|
quantized_activation_ops_test.cc
|
Sort header inclusions; define EIGEN_USE_THREADS where headers depend on it.
|
2017-06-07 07:41:38 -07:00 |
|
quantized_activation_ops.cc
|
Arm32/64 kernel optimizations:
|
2016-10-27 15:32:39 -07:00 |
|
quantized_add_op_test.cc
|
|
|
|
quantized_add_op.cc
|
|
|
|
quantized_batch_norm_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
quantized_batch_norm_op.cc
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
quantized_bias_add_op_test.cc
|
|
|
|
quantized_bias_add_op.cc
|
|
|
|
quantized_concat_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
quantized_concat_op.cc
|
Clean up OpKernel 1.
|
2020-01-29 15:49:57 -08:00 |
|
quantized_conv_ops_test.cc
|
Sort header inclusions; define EIGEN_USE_THREADS where headers depend on it.
|
2017-06-07 07:41:38 -07:00 |
|
quantized_conv_ops.cc
|
Splitting out common_shape_fns.cc:
|
2020-03-03 14:28:49 -08:00 |
|
quantized_instance_norm_test.cc
|
Internal change
|
2020-03-05 11:36:56 -08:00 |
|
quantized_instance_norm.cc
|
|
|
|
quantized_matmul_op_test.cc
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
quantized_matmul_op.cc
|
[TF:XLA] Enable more TF python tests for XLA:GPU
|
2019-05-24 00:40:13 -07:00 |
|
quantized_mul_op_test.cc
|
Internal change
|
2020-03-05 11:36:56 -08:00 |
|
quantized_mul_op.cc
|
Switch tensorflow::bit_cast -> absl::bit_cast.
|
2018-10-29 18:37:33 -07:00 |
|
quantized_pooling_ops_test.cc
|
Sort header inclusions; define EIGEN_USE_THREADS where headers depend on it.
|
2017-06-07 07:41:38 -07:00 |
|
quantized_pooling_ops.cc
|
supporting quantized pooling op
|
2020-12-01 09:08:37 -08:00 |
|
quantized_reshape_op_test.cc
|
Remove unused protobuf header inclusions
|
2017-06-06 03:15:27 -07:00 |
|
quantized_reshape_op.cc
|
Change quantize_graph in eightbit mode to remove FakeQuantWithMinMaxVars
|
2016-11-01 15:18:15 -07:00 |
|
quantized_resize_bilinear_op_test.cc
|
[Build cleanup] Split "core_cpu_impl" into fine-grained targets (4/n).
|
2020-04-28 09:53:59 -07:00 |
|
quantized_resize_bilinear_op.cc
|
Break up tensorflow/core/kernels/BUILD (part 3 of N):
|
2020-08-11 11:04:52 -07:00 |
|
queue_base.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
queue_base.h
|
Change some const declarations to constexpr
|
2020-04-20 14:59:17 -07:00 |
|
queue_op.cc
|
[TF:XLA] Add partial implementation of tf.FIFOQueue for XLA devices (e.g., TPU).
|
2018-06-28 20:26:49 -07:00 |
|
queue_op.h
|
[TF:XLA] Add partial implementation of tf.FIFOQueue for XLA devices (e.g., TPU).
|
2018-06-28 20:26:49 -07:00 |
|
queue_ops.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
ragged_cross_op.cc
|
Add tf.ragged.cross and tf.ragged.cross_hashed operations. These ops generate feature crosses from a list of input tensors (similar to tf.sparse.cross), and return the result as a RaggedTensor. Inputs may be tf.RaggedTensors, tf.Tensors, or tf.SparseTensors.
|
2020-03-02 18:13:45 -08:00 |
|
ragged_gather_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
ragged_gather_op.cc
|
Rollforward: Add uint32 & uint64 to TF_CALL_INTEGRAL_TYPES
|
2020-06-22 23:09:59 -07:00 |
|
ragged_range_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
ragged_range_op.cc
|
Update RaggedTensors to support int32 row_splits.
|
2019-04-24 19:39:31 -07:00 |
|
ragged_tensor_from_variant_op_test.cc
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
ragged_tensor_from_variant_op.cc
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
ragged_tensor_to_sparse_kernel_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
ragged_tensor_to_sparse_kernel.cc
|
Update RaggedTensors to support int32 row_splits.
|
2019-04-24 19:39:31 -07:00 |
|
ragged_tensor_to_tensor_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
ragged_tensor_to_tensor_op.cc
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
ragged_tensor_to_variant_op_test.cc
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
ragged_tensor_to_variant_op.cc
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
ragged_tensor_variant.cc
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
ragged_tensor_variant.h
|
Added gradients for RaggedTensorToVariant and RaggedTensorFromVariant. (This allows gradients to pass through map_fn when it is applied to ragged tensors.)
|
2020-10-14 10:09:39 -07:00 |
|
random_binomial_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 10:15:29 -07:00 |
|
random_binomial_op.cc
|
Remove duplicate definition of dcount.
|
2020-09-30 14:59:53 -07:00 |
|
random_binomial_op.h
|
|
|
|
random_op_cpu.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
random_op_gpu.cu.cc
|
|
|
|
random_op_gpu.h
|
fix offset datatype for philox
|
2021-01-14 12:51:24 -08:00 |
|
random_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
random_op.cc
|
Prevent int64 to int truncation in Shard API usage.
|
2020-09-18 17:34:10 -07:00 |
|
random_op.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
random_ops_util.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
random_poisson_op_test.cc
|
|
|
|
random_poisson_op.cc
|
random_poisson(rate=+inf) should return +inf, not nan.
|
2020-10-07 15:45:04 -07:00 |
|
random_poisson_op.h
|
|
|
|
random_shuffle_op.cc
|
|
|
|
random_shuffle_queue_op.cc
|
|
|
|
range_sampler_test.cc
|
Remove (Mutable)ArraySlice implementation and alias them to absl::Span.
|
2018-08-30 11:29:03 -07:00 |
|
range_sampler.cc
|
Merge pull request #41755 from tg-at-google:wsign-compare-semi-final-core2
|
2020-07-29 16:10:27 -07:00 |
|
range_sampler.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
reader_ops.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
record_input_op.cc
|
Update kernels and related libs to use tstring.
|
2019-08-27 18:22:08 -07:00 |
|
record_yielder.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
record_yielder.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
reduce_join_op.cc
|
|
|
|
reduction_gpu_kernels.cu.h
|
PR #44471: [ROCm] Update to use ROCm 3.9 (when building TF with --config=rocm)
|
2020-11-10 00:57:26 -08:00 |
|
reduction_ops_all.cc
|
Adding ROCm support for reduction ops
|
2019-07-03 12:44:32 +00:00 |
|
reduction_ops_any.cc
|
Adding ROCm support for reduction ops
|
2019-07-03 12:44:32 +00:00 |
|
reduction_ops_common_gpu.h
|
Adding ROCm support for reduction ops
|
2019-07-03 12:44:32 +00:00 |
|
reduction_ops_common.cc
|
Require that axis arguments to reduction ops are unique.
|
2020-07-01 15:45:41 -07:00 |
|
reduction_ops_common.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reduction_ops_euclidean.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reduction_ops_gpu_bool.cu.cc
|
Adding ROCm support for reduction ops
|
2019-07-03 12:44:32 +00:00 |
|
reduction_ops_gpu_complex64.cu.cc
|
Add a Euclidean norm reduction kernel. This implements a fused sqrt(reduce_sum(x * conj(x))) kernels for CPU (using Eigen) and GPU (using CUB), which is more efficient than the composite implementation at the TF level. It will also be easier to avoid the issue of producing NaNs in the gradient at the origin.
|
2019-02-15 12:14:56 -08:00 |
|
reduction_ops_gpu_complex128.cu.cc
|
Add a Euclidean norm reduction kernel. This implements a fused sqrt(reduce_sum(x * conj(x))) kernels for CPU (using Eigen) and GPU (using CUB), which is more efficient than the composite implementation at the TF level. It will also be easier to avoid the issue of producing NaNs in the gradient at the origin.
|
2019-02-15 12:14:56 -08:00 |
|
reduction_ops_gpu_double.cu.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_gpu_float.cu.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_gpu_int.cu.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_half_mean_sum.cu.cc
|
Adding ROCm support for reduction ops
|
2019-07-03 12:44:32 +00:00 |
|
reduction_ops_half_prod_max_min.cu.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_max.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_mean.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reduction_ops_min.cc
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
reduction_ops_prod.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reduction_ops_sum.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reduction_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
reduction_ops.h
|
|
|
|
redux_functor.h
|
Fix fp16 FusedBatchNorm CPU crash if batch dimension is 1.
|
2020-06-19 10:38:53 -07:00 |
|
reference_gemm.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
regex_full_match_op.cc
|
Make RegexFullMatchOp and RegexReplaceOp cache RE2 objects.
|
2020-08-11 09:44:05 -07:00 |
|
regex_replace_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
regex_replace_op.cc
|
Make RegexFullMatchOp and RegexReplaceOp cache RE2 objects.
|
2020-08-11 09:44:05 -07:00 |
|
relu_op_functor.h
|
working around a known gcc/hcc interface kernel args passing bug
|
2020-02-14 21:37:30 +00:00 |
|
relu_op_gpu.cu.cc
|
Merge pull request #40552 from kaixih:relu_grad_vect_pr
|
2020-06-18 16:10:59 -07:00 |
|
relu_op.cc
|
Merge pull request #43553 from Intel-tensorflow:fix-3898
|
2020-09-27 15:59:12 -07:00 |
|
relu_op.h
|
working around a known gcc/hcc interface kernel args passing bug
|
2020-02-14 21:37:30 +00:00 |
|
remote_fused_graph_execute_op_test_utils.cc
|
|
|
|
remote_fused_graph_execute_op_test_utils.h
|
|
|
|
remote_fused_graph_execute_op_test.cc
|
|
|
|
remote_fused_graph_execute_op.cc
|
|
|
|
remote_fused_graph_execute_utils_test.cc
|
|
|
|
remote_fused_graph_execute_utils.cc
|
tensorflow/core resolutions, set 2 of 3
|
2020-07-26 23:56:14 +00:00 |
|
remote_fused_graph_execute_utils.h
|
|
|
|
remote_fused_graph_rewriter_transform_test.cc
|
Use absl instead of deprecated str_util
|
2019-05-21 08:10:10 -07:00 |
|
remote_fused_graph_rewriter_transform.cc
|
Remove SplitAndParseAsInts.
|
2019-08-12 11:27:47 -07:00 |
|
requantization_range_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
requantization_range_op.cc
|
|
|
|
requantize_op_test.cc
|
Remove unused protobuf header inclusions
|
2017-06-06 03:15:27 -07:00 |
|
requantize.cc
|
Merge pull request #26229 from Intel-tensorflow:requantize_support_int8
|
2019-03-20 12:38:39 -07:00 |
|
reshape_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reshape_op.h
|
Clean up OpKernel 1.
|
2020-01-29 15:49:57 -08:00 |
|
reshape_util.cc
|
Rename tensorflow::Reshape() to tensorflow::ReshapeSparseTensor().
|
2020-04-08 10:47:41 -07:00 |
|
reshape_util.h
|
Rename tensorflow::Reshape() to tensorflow::ReshapeSparseTensor().
|
2020-04-08 10:47:41 -07:00 |
|
resource_variable_ops.cc
|
Register VariableShapeOp for TPU.
|
2021-02-08 15:50:16 -08:00 |
|
resource_variable_ops.h
|
Register VariableShapeOp for TPU.
|
2021-02-08 15:50:16 -08:00 |
|
restore_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
restore_op.cc
|
Expand all saveable operations to generate a single C++ restore call.
|
2018-01-04 12:06:05 -08:00 |
|
restore_v2_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
reverse_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
reverse_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
reverse_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
reverse_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
reverse_sequence_op_gpu.cu.cc
|
Adding ROCm support for the "reverse_sequence" op
|
2019-05-06 13:29:53 +00:00 |
|
reverse_sequence_op.cc
|
Fix argument check tests to work in eager mode
|
2020-05-18 15:53:45 -07:00 |
|
reverse_sequence_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
roll_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
roll_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
roll_op.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
roll_op.h
|
Merge pull request #27389 from musikisomorphie:roll_cuda
|
2019-06-27 20:26:22 -07:00 |
|
save_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
save_op.cc
|
Clean up OpKernel 1.
|
2020-01-29 15:49:57 -08:00 |
|
save_restore_tensor.cc
|
Use double for average computation.
|
2020-07-06 16:24:55 -07:00 |
|
save_restore_tensor.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
save_restore_v2_ops.cc
|
Use double for average computation.
|
2020-07-06 16:24:55 -07:00 |
|
save_v2_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
scan_ops_gpu_double.cu.cc
|
|
|
|
scan_ops_gpu_float.cu.cc
|
Implement a cumulative log-sum-exp operation.
|
2019-07-22 06:37:37 -07:00 |
|
scan_ops_gpu_half.cu.cc
|
Implement a cumulative log-sum-exp operation.
|
2019-07-22 06:37:37 -07:00 |
|
scan_ops_gpu_int.cu.cc
|
[ROCm] Fix for the broken --config=rocm build
|
2019-06-18 14:57:48 +00:00 |
|
scan_ops_gpu.h
|
PR #44471: [ROCm] Update to use ROCm 3.9 (when building TF with --config=rocm)
|
2020-11-10 00:57:26 -08:00 |
|
scan_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
scan_ops.cc
|
Implement a cumulative log-sum-exp operation.
|
2019-07-22 06:37:37 -07:00 |
|
scan_ops.h
|
Make tf.min, tf.max always propagate NaNs. Previously these ops followed the semantics of std::min and std:max and forwarded the first argument if either was NaN.
|
2020-10-20 21:29:20 -07:00 |
|
scatter_functor_gpu.cu.cc
|
Enable half for resource scatter
|
2020-07-10 20:36:44 -07:00 |
|
scatter_functor_gpu.cu.h
|
fix scatter op data type
|
2021-01-14 16:56:46 -08:00 |
|
scatter_functor.cc
|
Enable half for resource scatter
|
2020-07-10 20:36:44 -07:00 |
|
scatter_functor.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
scatter_nd_op_cpu_impl_0.cc
|
Changes to scatter_nd ops
|
2016-11-08 16:12:19 -08:00 |
|
scatter_nd_op_cpu_impl_1.cc
|
Adding CPU kernels for tf.scatter_nd(), tf.scatter_nd_update(), tf.scatter_nd_add(), tf.scatter_nd_sub(), tf.scatter_nd_mul() and tf.scatter_nd_div() as well as gradient functions for tf.scatter_nd() and tf.gather_nd()
|
2016-11-02 17:10:10 -07:00 |
|
scatter_nd_op_cpu_impl_2.cc
|
Adding CPU kernels for tf.scatter_nd(), tf.scatter_nd_update(), tf.scatter_nd_add(), tf.scatter_nd_sub(), tf.scatter_nd_mul() and tf.scatter_nd_div() as well as gradient functions for tf.scatter_nd() and tf.gather_nd()
|
2016-11-02 17:10:10 -07:00 |
|
scatter_nd_op_cpu_impl_3.cc
|
Adding CPU kernels for tf.scatter_nd(), tf.scatter_nd_update(), tf.scatter_nd_add(), tf.scatter_nd_sub(), tf.scatter_nd_mul() and tf.scatter_nd_div() as well as gradient functions for tf.scatter_nd() and tf.gather_nd()
|
2016-11-02 17:10:10 -07:00 |
|
scatter_nd_op_cpu_impl_4.cc
|
Adding CPU kernels for tf.scatter_nd(), tf.scatter_nd_update(), tf.scatter_nd_add(), tf.scatter_nd_sub(), tf.scatter_nd_mul() and tf.scatter_nd_div() as well as gradient functions for tf.scatter_nd() and tf.gather_nd()
|
2016-11-02 17:10:10 -07:00 |
|
scatter_nd_op_cpu_impl_5.cc
|
Adding CPU kernels for tf.scatter_nd(), tf.scatter_nd_update(), tf.scatter_nd_add(), tf.scatter_nd_sub(), tf.scatter_nd_mul() and tf.scatter_nd_div() as well as gradient functions for tf.scatter_nd() and tf.gather_nd()
|
2016-11-02 17:10:10 -07:00 |
|
scatter_nd_op_cpu_impl_6.cc
|
Make gather_nd & scatter_nd handle rank 6 and 7 indices.
|
2017-12-04 01:00:36 -08:00 |
|
scatter_nd_op_cpu_impl_7.cc
|
Make gather_nd & scatter_nd handle rank 6 and 7 indices.
|
2017-12-04 01:00:36 -08:00 |
|
scatter_nd_op_cpu_impl.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
scatter_nd_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
scatter_nd_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
scatter_nd_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
scatter_nd_op.h
|
Add Registration for non-stateful scatter_nd_min and scatter_nd_max.
|
2020-06-03 11:05:54 -07:00 |
|
scatter_op_gpu.cu.cc
|
Adding ROCm support for scatter ops
|
2019-06-10 14:51:44 +00:00 |
|
scatter_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
scatter_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
scoped_allocator_ops_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
scoped_allocator_ops.cc
|
Annotate alignment bytes in scoped allocator as initialized.
|
2019-09-09 12:47:48 -07:00 |
|
sdca_internal.cc
|
Remove call to std::random_shuffle.
|
2019-02-01 09:45:49 -08:00 |
|
sdca_internal.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
sdca_ops_test.cc
|
Preallocate vector storage when the ultimate vector size is known in advance
|
2017-06-01 09:45:21 -07:00 |
|
sdca_ops.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
searchsorted_op_gpu.cu.cc
|
Fixed a crash when calling tf.searchsorted with zero-sized values.
|
2020-03-06 16:30:32 -08:00 |
|
searchsorted_op.cc
|
remove out of date TODO
|
2019-11-26 10:06:23 -08:00 |
|
searchsorted_op.h
|
Add searchsorted (ie lower/upper bound) op.
|
2018-09-20 12:44:24 -07:00 |
|
segment_reduction_ops_gpu.cu.cc
|
Merge pull request #44436 from firejq:patch-5
|
2020-11-21 07:27:48 -08:00 |
|
segment_reduction_ops_impl_1.cc
|
Merge pull request #44436 from firejq:patch-5
|
2020-11-21 07:27:48 -08:00 |
|
segment_reduction_ops_impl_2.cc
|
Merge pull request #44436 from firejq:patch-5
|
2020-11-21 07:27:48 -08:00 |
|
segment_reduction_ops_impl_3.cc
|
Rename Segment*ReductionF->*ReductionF
|
2020-11-15 23:56:00 +08:00 |
|
segment_reduction_ops_impl_4.cc
|
Rename Segment*ReductionF->*ReductionF
|
2020-11-15 23:56:00 +08:00 |
|
segment_reduction_ops_impl_5.cc
|
|
|
|
segment_reduction_ops_impl.h
|
Merge pull request #44436 from firejq:patch-5
|
2020-11-21 07:27:48 -08:00 |
|
segment_reduction_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
segment_reduction_ops.h
|
Rename Segment*ReductionF->*ReductionF
|
2020-11-15 23:56:00 +08:00 |
|
sendrecv_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
sendrecv_ops.cc
|
Use TraceMeOp and TraceMeEncode to hide TraceString encoding details
|
2020-07-24 10:59:15 -07:00 |
|
sendrecv_ops.h
|
Use TraceMeOp and TraceMeEncode to hide TraceString encoding details
|
2020-07-24 10:59:15 -07:00 |
|
sequence_ops_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
sequence_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
serialize_sparse_op.cc
|
[tf.data] Further optimizations for SerializeManySparseOp.
|
2020-01-23 09:00:57 -08:00 |
|
session_ops.cc
|
Prevent segfault in GetSessionHandle{,V2}.
|
2020-09-18 16:30:02 -07:00 |
|
set_kernels.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
shape_ops_test.cc
|
Update benchmarks to use newer API
|
2020-11-24 16:59:20 -08:00 |
|
shape_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
shape_ops.h
|
|
|
|
slice_op_cpu_impl_1.cc
|
Split slice_op into multiple files to improve compilation times for CPU.
|
2016-08-16 17:47:45 -07:00 |
|
slice_op_cpu_impl_2.cc
|
Split slice_op into multiple files to improve compilation times for CPU.
|
2016-08-16 17:47:45 -07:00 |
|
slice_op_cpu_impl_3.cc
|
|
|
|
slice_op_cpu_impl_4.cc
|
Split slice_op into multiple files to improve compilation times for CPU.
|
2016-08-16 17:47:45 -07:00 |
|
slice_op_cpu_impl_5.cc
|
Split slice_op into multiple files to improve compilation times for CPU.
|
2016-08-16 17:47:45 -07:00 |
|
slice_op_cpu_impl_6.cc
|
Split slice_op into multiple files to improve compilation times for CPU.
|
2016-08-16 17:47:45 -07:00 |
|
slice_op_cpu_impl_7.cc
|
Increase maximum tensor rank to 7 for slice, strided_slice, and tile ops.
|
2017-01-24 16:04:06 -08:00 |
|
slice_op_cpu_impl_8.cc
|
Increase the supported number of dimensions for slice, strided_slice and pad to 8
|
2019-08-22 16:54:18 -07:00 |
|
slice_op_cpu_impl.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
slice_op_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
slice_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
slice_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
slice_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
smooth-hinge-loss.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
snapshot_op_gpu.cu.cc
|
Adding ROCm support for the "snapshot" op
|
2019-05-06 13:47:15 +00:00 |
|
snapshot_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
snapshot_op.h
|
Adding ROCm support for the "snapshot" op
|
2019-05-06 13:47:15 +00:00 |
|
sobol_op.cc
|
Add tf.math.sobol_sample for generating Sobol sequences (CPU implementation only).
|
2019-12-19 03:49:17 -08:00 |
|
softmax_op_functor.h
|
|
|
|
softmax_op_gpu.cu.cc
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
softmax_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
softplus_op_gpu.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
softplus_op.cc
|
PR #36526: Removed NDIMS template arg from BinaryElementWiseOp::Operate
|
2020-02-26 20:38:14 -08:00 |
|
softplus_op.h
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
softsign_op_gpu.cu.cc
|
|
|
|
softsign_op.cc
|
PR #36526: Removed NDIMS template arg from BinaryElementWiseOp::Operate
|
2020-02-26 20:38:14 -08:00 |
|
softsign_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
spacetobatch_benchmark_test.cc
|
|
|
|
spacetobatch_functor_gpu.cu.cc
|
Merge pull request #31223 from ThisIsIsaac:restrict
|
2019-08-28 08:28:47 -07:00 |
|
spacetobatch_functor.cc
|
Branch 183429339 (#16469)
|
2018-01-26 13:32:16 -08:00 |
|
spacetobatch_functor.h
|
|
|
|
spacetobatch_op.cc
|
Adding ROCm support for "space_to_batch" op
|
2019-05-21 21:08:05 +00:00 |
|
spacetodepth_op_gpu.cu.cc
|
Revert PR #43178: Make fast builds work with MSVC
|
2020-10-05 10:30:22 -07:00 |
|
spacetodepth_op.cc
|
rephrase wording
|
2020-08-31 16:42:37 +02:00 |
|
spacetodepth_op.h
|
Implement NCHW_VECT_C support for tf.space_to_depth on GPU.
|
2017-10-11 11:00:54 -07:00 |
|
sparse_add_grad_op.cc
|
|
|
|
sparse_add_op_test.cc
|
|
|
|
sparse_add_op.cc
|
Fixing nondeterministic nullptr access.
|
2020-06-05 22:45:43 -07:00 |
|
sparse_concat_op.cc
|
|
|
|
sparse_conditional_accumulator_op.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
sparse_conditional_accumulator.h
|
|
|
|
sparse_cross_op.cc
|
Modify type-specific mappings to be explicit about long/long long rather than
|
2020-07-15 11:47:51 -07:00 |
|
sparse_dense_binary_op_shared_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
sparse_dense_binary_op_shared.cc
|
Use absl instead of deprecated str_util
|
2019-05-21 08:10:10 -07:00 |
|
sparse_fill_empty_rows_op.cc
|
Merge pull request #44719 from benbarsdell:gpu-SparseFillEmptyRows-cpu-refactor
|
2020-12-18 18:06:07 -08:00 |
|
sparse_fill_empty_rows_op.h
|
Rollback of: PR #44719: Refactor SparseFillEmptyRows[Grad] CPU kernels into classes
|
2020-12-03 15:15:13 -08:00 |
|
sparse_matmul_op_test.cc
|
|
|
|
sparse_matmul_op.cc
|
Merge pull request #40962 from Intel-tensorflow:yang/eigen-bf16
|
2020-07-14 17:21:38 -07:00 |
|
sparse_matmul_op.h
|
Inline windows/cpu_info.h into platform/cpu_info.h. Since the endianness macros are now provided by byte_order.h, we can remove all other inclusion sites.
|
2019-08-07 23:10:52 -07:00 |
|
sparse_reduce_op.cc
|
Fix 64-bit integer portability problems in TensorFlow kernels.
|
2020-01-21 19:14:20 -08:00 |
|
sparse_reduce_sum_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
sparse_reorder_op.cc
|
Merge pull request #45675 from yongtang:45392-tf.sparse.reorder
|
2020-12-30 12:42:27 -08:00 |
|
sparse_reshape_op.cc
|
Rename tensorflow::Reshape() to tensorflow::ReshapeSparseTensor().
|
2020-04-08 10:47:41 -07:00 |
|
sparse_slice_grad_op.cc
|
Use the safe sparse tensor API that returns errors rather than crashing
|
2018-07-16 12:10:09 -07:00 |
|
sparse_slice_op.cc
|
Use the safe sparse tensor API that returns errors rather than crashing
|
2018-07-16 12:10:09 -07:00 |
|
sparse_softmax_op.cc
|
|
|
|
sparse_sparse_binary_op_shared.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
sparse_split_op.cc
|
Add support for negative indices to sparse_split.
|
2020-07-07 16:37:22 -07:00 |
|
sparse_tensor_dense_add_op.cc
|
|
|
|
sparse_tensor_dense_add_op.h
|
|
|
|
sparse_tensor_dense_matmul_op_gpu.cu.cc
|
|
|
|
sparse_tensor_dense_matmul_op_test.cc
|
Update copyright for 3p/tf/core.
|
2016-06-02 13:39:22 -07:00 |
|
sparse_tensor_dense_matmul_op.cc
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
sparse_tensor_dense_matmul_op.h
|
|
|
|
sparse_tensors_map_ops.cc
|
|
|
|
sparse_to_dense_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 14:08:15 -07:00 |
|
sparse_to_dense_op.cc
|
Make sparse_to_dense_op_py_test v2 friendly
|
2020-07-10 16:05:14 -07:00 |
|
sparse_utils_test.cc
|
Making GetStartIndicesOfEachDenseRow produce leading zeros in case if several beginning empty rows are empty.
|
2020-04-22 01:30:52 -07:00 |
|
sparse_utils.cc
|
Making GetStartIndicesOfEachDenseRow produce leading zeros in case if several beginning empty rows are empty.
|
2020-04-22 01:30:52 -07:00 |
|
sparse_utils.h
|
Making GetStartIndicesOfEachDenseRow produce leading zeros in case if several beginning empty rows are empty.
|
2020-04-22 01:30:52 -07:00 |
|
sparse_xent_op_gpu.cu.cc
|
|
|
|
sparse_xent_op_test.cc
|
Rolling back PR #44894 because it caused type check errors (got bfloat16 when expecting float).
|
2020-12-07 17:30:30 -08:00 |
|
sparse_xent_op.cc
|
Rolling back PR #44894 because it caused type check errors (got bfloat16 when expecting float).
|
2020-12-07 17:30:30 -08:00 |
|
sparse_xent_op.h
|
Update references to bounds_checks.h
|
2019-01-22 14:29:28 -08:00 |
|
spectrogram_convert_test_data.cc
|
Add AudioSpectrogram op to TensorFlow for audio feature generation
|
2017-04-11 16:08:23 -07:00 |
|
spectrogram_op_test.cc
|
|
|
|
spectrogram_op.cc
|
fix support of multichaannel data in audio_spectrogram
|
2020-03-16 10:26:05 -07:00 |
|
spectrogram_test_utils.cc
|
Remove SplitAndParseAsFloats from tensorflow::strings API.
|
2019-08-11 00:56:38 -07:00 |
|
spectrogram_test_utils.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
spectrogram_test.cc
|
|
|
|
spectrogram.cc
|
|
|
|
spectrogram.h
|
|
|
|
split_lib_cpu.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
split_lib_gpu.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
split_lib_gpu.h
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
split_lib.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
split_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
split_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
split_v_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
split_v_op.cc
|
Merge pull request #40116 from kitstar:master
|
2020-07-22 10:54:12 -07:00 |
|
squared-loss.h
|
|
|
|
stack_ops.cc
|
Register default kernels for stack* ops.
|
2019-06-24 14:51:00 -07:00 |
|
stack.cc
|
Merge pull request #40697 from tg-at-google:sign-compare-warning-fixes-batch-3
|
2020-06-24 14:38:10 -07:00 |
|
stack.h
|
[TF:XLA] Register the standard Stack kernels on XLA_... devices.
|
2018-11-05 10:56:21 -08:00 |
|
stage_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
stateful_random_ops_cpu_gpu.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
stateful_random_ops_gpu.cu.cc
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
stateful_random_ops.cc
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
stateful_random_ops.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
stateless_random_gamma_op_gpu.cu.cc
|
Add GPU kernel for tf.random.stateless_gamma.
|
2021-01-15 03:01:47 -08:00 |
|
stateless_random_gamma_op.cc
|
Add GPU kernel for tf.random.stateless_gamma.
|
2021-01-15 03:01:47 -08:00 |
|
stateless_random_gamma_op.h
|
Add GPU kernel for tf.random.stateless_gamma.
|
2021-01-15 03:01:47 -08:00 |
|
stateless_random_ops_v2.cc
|
Splits StatelessRandomGetKeyCounter into two ops because XLA propagates a compile-time-constant requirement (originating from V2 stateless RNG ops) on output alg to input seed, which prevents seed from depending on a variable. With the split ops, alg no longer depends on seed.
|
2020-11-16 12:17:48 -08:00 |
|
stateless_random_ops_v2.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
stateless_random_ops.cc
|
Move tf.random.stateless_gamma CPU kernel to a separate file.
|
2020-12-23 10:22:47 -08:00 |
|
stateless_random_ops.h
|
|
|
|
strided_slice_op_define_grad.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_gpu_bool.cu.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_gpu_complex.cu.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
strided_slice_op_gpu_impl.h
|
Increase the supported number of dimensions for slice, strided_slice and pad to 8
|
2019-08-22 16:54:18 -07:00 |
|
strided_slice_op_gpu_int.cu.cc
|
Register uint8, int16, and uint16 for strided slice on GPU, such that it supports the same types as on CPU.
|
2020-05-27 15:35:39 -07:00 |
|
strided_slice_op_gpu_number_types.cu.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_impl.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
strided_slice_op_inst_0.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_1.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_2.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_3.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_4.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_5.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_6.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_7.cc
|
Adding ROCm support for slice and strided_slice ops
|
2019-06-10 15:20:12 +00:00 |
|
strided_slice_op_inst_8.cc
|
|
|
|
strided_slice_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
strided_slice_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
strided_slice_op.h
|
Fixes/updates for tensorflow::tstring migration.
|
2019-11-26 23:35:38 -08:00 |
|
string_format_op_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-23 18:18:03 -08:00 |
|
string_format_op.cc
|
Fix some unnecessary string copies when reading from or writing to a Tensor.
|
2019-08-08 15:38:48 -07:00 |
|
string_join_op.cc
|
Migrated a subset of kernels to use tstring.
|
2019-08-14 09:29:06 -07:00 |
|
string_length_op.cc
|
|
|
|
string_lower_op.cc
|
Change comment of the external repo #include file origin.
|
2020-03-23 10:42:53 -07:00 |
|
string_ngrams_op_test.cc
|
Update kernels and related libs to use tstring.
|
2019-08-27 18:22:08 -07:00 |
|
string_ngrams_op.cc
|
Validate data_splits for tf.StringNGrams.
|
2020-09-18 16:00:29 -07:00 |
|
string_split_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
string_split_op.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
string_strip_op.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
string_to_hash_bucket_fast_op.cc
|
Adds StringToHashBucketFast Op to TF android selective registration builds, by split out the fast op into its own .h/.cc
|
2020-11-17 17:36:31 -08:00 |
|
string_to_hash_bucket_fast_op.h
|
Adds StringToHashBucketFast Op to TF android selective registration builds, by split out the fast op into its own .h/.cc
|
2020-11-17 17:36:31 -08:00 |
|
string_to_hash_bucket_op.cc
|
Adds StringToHashBucketFast Op to TF android selective registration builds, by split out the fast op into its own .h/.cc
|
2020-11-17 17:36:31 -08:00 |
|
string_to_hash_bucket_op.h
|
Adds StringToHashBucketFast Op to TF android selective registration builds, by split out the fast op into its own .h/.cc
|
2020-11-17 17:36:31 -08:00 |
|
string_to_number_op.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
string_upper_op.cc
|
Change comment of the external repo #include file origin.
|
2020-03-23 10:42:53 -07:00 |
|
string_util.cc
|
Add unicode_transcode Op to TF strings package
|
2018-10-25 14:10:48 -07:00 |
|
string_util.h
|
Add unicode_transcode Op to TF strings package
|
2018-10-25 14:10:48 -07:00 |
|
substr_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
substr_op.cc
|
Start llvm-bazel integration
|
2021-02-09 10:00:38 -08:00 |
|
summary_audio_op_test.cc
|
Migrated a subset of kernels to use tstring.
|
2019-08-14 09:29:06 -07:00 |
|
summary_audio_op.cc
|
Clean up OpKernel 1.
|
2020-01-29 15:49:57 -08:00 |
|
summary_image_op_test.cc
|
|
|
|
summary_image_op.cc
|
|
|
|
summary_interface.h
|
|
|
|
summary_kernels.cc
|
Updated the majority of string tensor accessors to use tstring type.
|
2019-08-07 11:46:35 -07:00 |
|
summary_op_test.cc
|
removed scalar summary tests from tensorflow/core
|
2020-07-27 23:31:57 +00:00 |
|
summary_tensor_op_test.cc
|
Migrated a subset of kernels to use tstring.
|
2019-08-14 09:29:06 -07:00 |
|
summary_tensor_op.cc
|
Use ParseFromTString/ParseFromArray for tstring.
|
2020-02-04 14:18:15 -08:00 |
|
tensor_array_ops.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
tensor_array.cc
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
tensor_array.h
|
Merge various kernel registrations with macros
|
2020-06-08 00:42:40 -07:00 |
|
tensor_cord_test.cc
|
Fixes/updates for tensorflow::tstring migration.
|
2019-11-26 23:35:38 -08:00 |
|
tensor_cord.cc
|
TensorCord Implementation.
|
2019-10-25 16:02:26 -07:00 |
|
tensor_cord.h
|
Avoid designated initializers, as MSVC complains about them.
|
2020-01-16 09:58:55 -08:00 |
|
tensor_flag_utils_test.cc
|
Fix 64-bit integer portability problems in TensorFlow kernels.
|
2020-01-21 19:14:20 -08:00 |
|
tensor_flag_utils.cc
|
Add some utils for parsing Tensors as simple functions at runtime.
|
2019-01-18 09:59:50 -08:00 |
|
tensor_flag_utils.h
|
PR #26235: typos in tensorflow/core fixed
|
2019-03-08 14:50:45 -08:00 |
|
tensor_list.cc
|
Define TensorList class in a separate library
|
2020-01-24 17:37:07 -08:00 |
|
tensor_list.h
|
Fix several narrowing issues on 32-bit devices
|
2020-08-11 15:48:13 -07:00 |
|
tensor_map_test.cc
|
Merge pull request #42300 from kttian:list_keys
|
2020-08-25 10:12:32 -07:00 |
|
tensor_map.cc
|
Merge pull request #42141 from kttian:erase_grad
|
2020-08-12 17:42:08 -07:00 |
|
tensor_map.h
|
Merge pull request #42300 from kttian:list_keys
|
2020-08-25 10:12:32 -07:00 |
|
text_line_reader_op.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
tf_record_reader_op.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
tile_functor_cpu_bfloat16.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_bool.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_complex64.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_complex128.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_double.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_float.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_half.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_int8.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_int16.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_int32.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_int64.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_tstring.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_uint8.cc
|
[Build time] Split "tile_functor_cpu.cc" into a shared header and type-specific .cc files.
|
2020-04-02 17:09:49 -07:00 |
|
tile_functor_cpu_uint32.cc
|
Add uint32/uint64 support for tf.tile
|
2020-05-11 17:46:29 +00:00 |
|
tile_functor_cpu_uint64.cc
|
Add uint32/uint64 support for tf.tile
|
2020-05-11 17:46:29 +00:00 |
|
tile_functor_cpu_variant.cc
|
Allow tile op to work on variant dtype.
|
2020-05-19 16:55:56 -07:00 |
|
tile_functor_cpu.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
tile_functor_gpu_bool.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_complex64.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_complex128.cu.cc
|
|
|
|
tile_functor_gpu_double.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_float.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_half.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_int16.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_int32.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu_int64.cu.cc
|
Adding ROCm support for tile ops
|
2019-05-20 18:18:42 +00:00 |
|
tile_functor_gpu.h
|
Fix stack-buffer-overflow in tile_functor_gpu.h. host_buf has size = size(int32)*num elements but we are trying to copy size(int64)*num elements out of it.
|
2020-10-22 14:40:40 -07:00 |
|
tile_functor_sycl.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
tile_functor.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
tile_ops_cpu_impl_1.cc
|
Shard tile_ops.cc for cpu, to reduce compilation times.
|
2016-08-16 10:01:57 -07:00 |
|
tile_ops_cpu_impl_2.cc
|
Shard tile_ops.cc for cpu, to reduce compilation times.
|
2016-08-16 10:01:57 -07:00 |
|
tile_ops_cpu_impl_3.cc
|
Shard tile_ops.cc for cpu, to reduce compilation times.
|
2016-08-16 10:01:57 -07:00 |
|
tile_ops_cpu_impl_4.cc
|
Shard tile_ops.cc for cpu, to reduce compilation times.
|
2016-08-16 10:01:57 -07:00 |
|
tile_ops_cpu_impl_5.cc
|
Shard tile_ops.cc for cpu, to reduce compilation times.
|
2016-08-16 10:01:57 -07:00 |
|
tile_ops_cpu_impl_6.cc
|
Automated rollback of change 130659785
|
2016-08-18 13:47:37 -07:00 |
|
tile_ops_cpu_impl_7.cc
|
Increase maximum tensor rank to 7 for slice, strided_slice, and tile ops.
|
2017-01-24 16:04:06 -08:00 |
|
tile_ops_cpu_impl.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
tile_ops_gpu_impl_1.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_2.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_3.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_4.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_5.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_6.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_7.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl_8.cu.cc
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
tile_ops_gpu_impl.h
|
|
|
|
tile_ops_impl.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
tile_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
topk_op_gpu_double.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_float.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_half.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_int8.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_int16.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_int32.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_int64.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_uint8.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_uint16.cu.cc
|
Fixing and enabling TopK on ROCm
|
2020-09-16 16:05:37 +00:00 |
|
topk_op_gpu_uint32.cu.cc
|
Define ROCM TopKFunctor for uint{32,64}.
|
2020-10-20 16:27:58 -07:00 |
|
topk_op_gpu_uint64.cu.cc
|
Define ROCM TopKFunctor for uint{32,64}.
|
2020-10-20 16:27:58 -07:00 |
|
topk_op_gpu.h
|
Fix ROCM build after #39429 and ce2a22c1dd
|
2020-10-19 17:06:19 -07:00 |
|
topk_op.cc
|
Merge pull request #39429 from ekuznetsov139:google-upstream-topk
|
2020-10-18 22:12:36 -07:00 |
|
topk_op.h
|
fix C++ header guards.
|
2018-08-21 16:22:05 -07:00 |
|
training_op_helpers.cc
|
Changing the copy-on-write semantics of resource variables.
|
2018-12-10 12:18:13 -08:00 |
|
training_op_helpers.h
|
Use RefCountPtr in LookupResource to avoid leaks
|
2019-05-28 22:55:44 -07:00 |
|
training_ops_gpu.cu.cc
|
Avoid "too many resources requested for launch" errors for SparseApplyFtrl
|
2021-02-01 11:34:11 -08:00 |
|
training_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
training_ops.cc
|
Add GPU kernels for SparseApply[Proximal]Adagrad
|
2020-12-18 10:52:39 +11:00 |
|
training_ops.h
|
Address minor review comments from PR 44919
|
2020-11-23 18:14:03 +11:00 |
|
transpose_functor_cpu.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
transpose_functor_gpu.cu.cc
|
Fix size computation logic in TransposeSimple
|
2020-07-14 11:36:38 -07:00 |
|
transpose_functor.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
transpose_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
transpose_op.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
transpose_util_test.cc
|
|
|
|
typed_conditional_accumulator_base.h
|
|
|
|
typed_queue.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
unary_ops_composition_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
unary_ops_composition.cc
|
Change some const declarations to constexpr
|
2020-04-20 14:59:17 -07:00 |
|
unicode_ops.cc
|
Change comment of the external repo #include file origin.
|
2020-03-23 10:42:53 -07:00 |
|
unicode_script_op.cc
|
|
|
|
unique_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
unique_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
unpack_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
unravel_index_op.cc
|
Fix several narrowing issues on 32-bit devices
|
2020-08-11 15:48:13 -07:00 |
|
unsorted_segment_join_op.cc
|
|
|
|
variable_ops_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
variable_ops.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
variable_ops.h
|
Optimize (legacy, reference-typed) VariableOp::Compute().
|
2020-03-27 14:48:18 -07:00 |
|
where_op_gpu_impl_1.cu.cc
|
Automated g4 rollback of changelist 170772848
|
2017-10-10 10:49:14 -07:00 |
|
where_op_gpu_impl_2.cu.cc
|
Automated g4 rollback of changelist 170772848
|
2017-10-10 10:49:14 -07:00 |
|
where_op_gpu_impl_3.cu.cc
|
Automated g4 rollback of changelist 170772848
|
2017-10-10 10:49:14 -07:00 |
|
where_op_gpu_impl_4.cu.cc
|
Automated g4 rollback of changelist 170772848
|
2017-10-10 10:49:14 -07:00 |
|
where_op_gpu_impl_5.cu.cc
|
Automated g4 rollback of changelist 170772848
|
2017-10-10 10:49:14 -07:00 |
|
where_op_gpu_impl_6.cu.cc
|
Allow tf.compat.v2.where to support Tensors with rank up to 8.
|
2019-06-20 17:31:07 -07:00 |
|
where_op_gpu_impl_7.cu.cc
|
Allow tf.compat.v2.where to support Tensors with rank up to 8.
|
2019-06-20 17:31:07 -07:00 |
|
where_op_gpu_impl_8.cu.cc
|
Allow tf.compat.v2.where to support Tensors with rank up to 8.
|
2019-06-20 17:31:07 -07:00 |
|
where_op_gpu.cu.h
|
Merge pull request #36640 from ROCmSoftwarePlatform:google-upstream-gpuprim
|
2020-04-15 09:18:52 -07:00 |
|
where_op.cc
|
fix typos in core directory
|
2020-10-29 02:52:55 +03:00 |
|
where_op.h
|
[WhereCPUOp] Optimize the tensor-wrapping of the num_true scalar.
|
2020-02-24 11:00:44 -08:00 |
|
whole_file_read_ops.cc
|
Updated ReaderInterface and subclasses to use tstring.
|
2019-08-27 20:01:38 -07:00 |
|
winograd_transform.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
|
word2vec_kernels.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
xent_op_gpu.cu.cc
|
Check if GOOGLE_CUDA and TENSORFLOW_USE_ROCM are defined before using them.
|
2019-04-24 17:19:21 -07:00 |
|
xent_op_test.cc
|
Internal tests cleanup.
|
2020-10-29 15:56:05 -07:00 |
|
xent_op.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
xent_op.h
|
Fix include guards
|
2018-08-20 11:20:05 -07:00 |
|
xsmm_conv2d_test.cc
|
Merge pull request #24706 from hfp:upstream
|
2019-11-19 12:31:39 -08:00 |
|
xsmm_conv2d.cc
|
Merge pull request #24706 from hfp:upstream
|
2019-11-19 12:31:39 -08:00 |
|
xsmm_conv2d.h
|
Remove THIRD_PARTY_ from #include guards
|
2018-01-24 14:31:28 -08:00 |
