| .. |
|
allocation_description.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
allocator_registry.cc
|
in resolution of [Wsign-compare] warning id 9
|
2020-06-02 11:07:21 -04:00 |
|
allocator_registry.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
allocator_test.cc
|
Internal tests cleanup.
|
2020-10-29 23:45:48 -07:00 |
|
allocator.cc
|
Replace a couple of Enable$FOO(bool) patterns with an Enable and a Disable function; NFC
|
2020-08-14 11:35:36 -07:00 |
|
allocator.h
|
Support coalescing of adjacent allocation regions in BFC allocator.
|
2021-01-13 15:47:59 -08:00 |
|
api_def.proto
|
Merge pull request #36387 from jhseu:go_fix
|
2020-02-03 17:45:49 -08:00 |
|
attr_value_util_test.cc
|
Update kMaxListSummarySize to 50
|
2020-07-22 10:28:31 -07:00 |
|
attr_value_util.cc
|
Update kMaxListSummarySize to 50
|
2020-07-22 10:28:31 -07:00 |
|
attr_value_util.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
attr_value.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
batch_util_test.cc
|
add a tensorflow::batch_util::CopyContiguousSlices utility function for
|
2020-05-27 10:26:55 -07:00 |
|
bfloat16_test.cc
|
Internal tests cleanup.
|
2020-10-29 23:45:48 -07:00 |
|
bfloat16.cc
|
Get rid of undefined behavior in bfloat16 <-> float conversion routines by using vectorized Eigen expressions or memcpy.
|
2020-08-16 14:31:02 -07:00 |
|
bfloat16.h
|
Get rid of undefined behavior in bfloat16 <-> float conversion routines by using vectorized Eigen expressions or memcpy.
|
2020-08-16 14:31:02 -07:00 |
|
bounds_check.h
|
|
|
|
BUILD
|
[tf.data] Add a new proto to store tf.data.Options data and conversion functions between tf.data.Options and their proto representation.
|
2021-02-05 15:55:03 -08:00 |
|
cancellation_test.cc
|
|
|
|
cancellation.cc
|
[tf.data] Update autotuning implementation to increase optimization frequency any time the input pipeline graph changes.
|
2021-02-03 15:48:24 -08:00 |
|
cancellation.h
|
[tf.data] Update autotuning implementation to increase optimization frequency any time the input pipeline graph changes.
|
2021-02-03 15:48:24 -08:00 |
|
collective.cc
|
Ensure that CollectiveParams outlives all references to it.
|
2021-02-04 10:07:20 -08:00 |
|
collective.h
|
Ensure that CollectiveParams outlives all references to it.
|
2021-02-04 10:07:20 -08:00 |
|
common_shape_fns_test.cc
|
Support computing exponential running mean and variance in fused_batch_norm.
|
2020-02-26 14:31:32 -08:00 |
|
common_shape_fns.cc
|
replace alternative operator to fix msvc build break
|
2021-01-24 23:01:48 +08:00 |
|
common_shape_fns.h
|
Resolved merge conflicts
|
2020-10-09 08:42:49 -07:00 |
|
control_flow.h
|
|
|
|
cost_graph.proto
|
Add aggregated cost dimensions to the CostGraph, typically used for
|
2020-03-05 13:48:47 -08:00 |
|
cpu_allocator_impl.cc
|
Support coalescing of adjacent allocation regions in BFC allocator.
|
2021-01-13 15:47:59 -08:00 |
|
dataset_options.proto
|
[tf.data] Add a new proto to store tf.data.Options data and conversion functions between tf.data.Options and their proto representation.
|
2021-02-05 15:55:03 -08:00 |
|
dataset_stateful_op_allowlist.h
|
Rename dataset_stateful_op_whitelist to dataset_stateful_op_allowlist
|
2020-07-30 10:36:40 -07:00 |
|
dataset_test.cc
|
[tf.data] Implement splitting for tensor_slices_dataset.
|
2020-09-16 14:05:56 -07:00 |
|
dataset.cc
|
[tf.data] Include information about element shapes and types in TraceMe metadata.
|
2021-02-18 14:04:51 -08:00 |
|
dataset.h
|
[tf.data] Include information about element shapes and types in TraceMe metadata.
|
2021-02-18 14:04:51 -08:00 |
|
device_attributes.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
device_base_test.cc
|
|
|
|
device_base.cc
|
replace unreachable returns by aborts
|
2020-08-28 18:26:11 +02:00 |
|
device_base.h
|
Add stream executor c api extension
|
2020-11-11 08:51:08 +00:00 |
|
device_factory.cc
|
Move Device and DeviceFactory to core/framework. https://github.com/tensorflow/tensorflow/pull/43611/ needs to access device registry to get a subtype for a given device type as a part of kernel registration. Kernel registration is a part of framework and should not depend on common_runtime to avoid a circular dependency.
|
2020-10-16 15:11:35 -07:00 |
|
device_factory.h
|
Move Device and DeviceFactory to core/framework. https://github.com/tensorflow/tensorflow/pull/43611/ needs to access device registry to get a subtype for a given device type as a part of kernel registration. Kernel registration is a part of framework and should not depend on common_runtime to avoid a circular dependency.
|
2020-10-16 15:11:35 -07:00 |
|
device.cc
|
Merge pull request #43812 from ppiskorski:get_device_context
|
2020-10-23 18:14:22 -07:00 |
|
device.h
|
replace PFLR DeviceGetContext hardcode with Device::IsRemoteCallAllowed
|
2020-10-22 20:49:03 +02:00 |
|
extension_type_variant_test.cc
|
Add ExtensionTypeVariant class, which can be used to store any ExtensionType values (aka CompositeTensor values) in a variant scalar.
|
2020-11-23 10:00:58 -08:00 |
|
extension_type_variant.cc
|
Add ExtensionTypeVariant class, which can be used to store any ExtensionType values (aka CompositeTensor values) in a variant scalar.
|
2020-11-23 10:00:58 -08:00 |
|
extension_type_variant.h
|
Add ExtensionTypeVariant class, which can be used to store any ExtensionType values (aka CompositeTensor values) in a variant scalar.
|
2020-11-23 10:00:58 -08:00 |
|
fake_input.cc
|
|
|
|
fake_input.h
|
|
|
|
function_handle_cache.cc
|
This CL optimizes C++11 range-based for loops where the variable is copied in each iteration but it would suffice to obtain it by const reference. This is only applied to loop variables of types that are expensive to copy which means they are not trivially copyable or have a non-trivial copy constructor or destructor.
|
2020-04-07 14:38:02 -07:00 |
|
function_handle_cache.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
function_test.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 |
|
function_testlib.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 |
|
function_testlib.h
|
|
|
|
function.cc
|
Copy stack traces to instantiated functions.
|
2020-12-08 16:33:18 -08:00 |
|
function.h
|
[NFC] Optimize getting the last stack frame + stack frame conversion
|
2020-12-10 22:55:07 -08:00 |
|
function.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
graph_def_util_test.cc
|
|
|
|
graph_def_util.cc
|
|
|
|
graph_def_util.h
|
|
|
|
graph_to_functiondef_test.cc
|
[Build cleanup] Update #includes of moved header "graph/graph_constructor.h".
|
2020-04-29 09:20:48 -07:00 |
|
graph_to_functiondef.cc
|
Ensure that shapes are set for captured inputs to functions. Small eager tensors are captured as const ops, which don't have a shape attr.
|
2020-09-21 14:07:28 -07:00 |
|
graph_to_functiondef.h
|
NFC: Refactored existing runtime code to prepare for TFRT/TF2 training's integration points.
|
2020-12-15 13:26:34 -08:00 |
|
graph_transfer_info.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
graph.proto
|
Remove outdated comment :P
|
2020-11-16 11:38:56 -08:00 |
|
kernel_def_builder_test.cc
|
Use explicitly brace-initialized int64 values rather than LL suffixes to protect against situations where int64 is not long long.
|
2020-07-13 17:02:18 -07:00 |
|
kernel_def_builder.cc
|
|
|
|
kernel_def_builder.h
|
Delete stray commented line
|
2020-11-03 13:39:20 -08:00 |
|
kernel_def_util_test.cc
|
|
|
|
kernel_def_util.cc
|
|
|
|
kernel_def_util.h
|
|
|
|
kernel_def.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
kernel_shape_util.cc
|
Splitting out common_shape_fns.cc:
|
2020-03-03 14:28:49 -08:00 |
|
kernel_shape_util.h
|
Splitting out common_shape_fns.cc:
|
2020-03-03 14:28:49 -08:00 |
|
load_library.cc
|
Rename LoadLibrary to avoid conflict with Windows macros
|
2020-08-04 19:24:11 -07:00 |
|
local_rendezvous.cc
|
Add Ref/Unref to make sure rendezvous outlives its cancellation callback.
|
2020-10-02 11:23:42 -07:00 |
|
local_rendezvous.h
|
Add Ref/Unref to make sure rendezvous outlives its cancellation callback.
|
2020-10-02 11:23:42 -07:00 |
|
log_memory.cc
|
Optimize calls to std::string::find() and friends for a single char.
|
2020-12-17 18:11:13 -08:00 |
|
log_memory.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
log_memory.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
logging.cc
|
|
|
|
logging.h
|
|
|
|
lookup_interface.cc
|
Mutablehashtable lookup support full size dynamic default values.
|
2020-10-26 19:05:22 +08:00 |
|
lookup_interface.h
|
Merge pull request #43269 from rhdong:RFC237-patch-for-recsys-sig
|
2020-10-27 08:06:45 -07:00 |
|
memory_types_test.cc
|
Extend MemoryTypesForNode test
|
2020-08-06 23:44:20 +02:00 |
|
memory_types.cc
|
Set the default value for the use_gpu-param to True in test utilities
|
2021-02-10 22:58:39 -08:00 |
|
memory_types.h
|
|
|
|
metrics.cc
|
[tf.data] Add metric to record file names read by tf.data Datasets.
|
2021-02-05 11:24:53 -08:00 |
|
metrics.h
|
[tf.data] Add metric to record file names read by tf.data Datasets.
|
2021-02-05 11:24:53 -08:00 |
|
model_test.cc
|
Introducing functionality to save and load an autotuning model and its optimization parameters to/from a file. Model saving happens in a background thread (separate from the optimization thread) after each optimization run. It is activated by setting an environment variable AUTOTUNE_DEBUG_DIR.
|
2021-02-17 13:34:31 -08:00 |
|
model.cc
|
Introducing functionality to save and load an autotuning model and its optimization parameters to/from a file. Model saving happens in a background thread (separate from the optimization thread) after each optimization run. It is activated by setting an environment variable AUTOTUNE_DEBUG_DIR.
|
2021-02-17 13:34:31 -08:00 |
|
model.h
|
Introducing functionality to save and load an autotuning model and its optimization parameters to/from a file. Model saving happens in a background thread (separate from the optimization thread) after each optimization run. It is activated by setting an environment variable AUTOTUNE_DEBUG_DIR.
|
2021-02-17 13:34:31 -08:00 |
|
model.proto
|
Introducing functionality to save and load an autotuning model and its optimization parameters to/from a file. Model saving happens in a background thread (separate from the optimization thread) after each optimization run. It is activated by setting an environment variable AUTOTUNE_DEBUG_DIR.
|
2021-02-17 13:34:31 -08:00 |
|
node_def_builder_test.cc
|
|
|
|
node_def_builder.cc
|
Removed USE_TSTRING usage.
|
2020-01-28 22:26:26 -08:00 |
|
node_def_builder.h
|
Removed USE_TSTRING usage.
|
2020-01-28 22:26:26 -08:00 |
|
node_def_util_test.cc
|
Addressing large number of outputs in a consistent way across our various C
|
2020-10-26 17:18:34 -07:00 |
|
node_def_util.cc
|
Addressing large number of outputs in a consistent way across our various C
|
2020-10-26 17:18:34 -07:00 |
|
node_def_util.h
|
Fix colocation in function inlining
|
2020-09-17 13:43:16 -07:00 |
|
node_def.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
node_properties_test.cc
|
Replace NodeDef with std::shared_ptr<NodeProperties> in the kernel creation code paths and try to avoid as many copies of NodeDefs as possible. This will in most cases allow sharing the NodeDef between the OpKernel and the graph Node from which it is created.
|
2020-02-18 13:20:06 -08:00 |
|
node_properties.cc
|
Replace NodeDef with std::shared_ptr<NodeProperties> in the kernel creation code paths and try to avoid as many copies of NodeDefs as possible. This will in most cases allow sharing the NodeDef between the OpKernel and the graph Node from which it is created.
|
2020-02-18 13:20:06 -08:00 |
|
node_properties.h
|
Replace NodeDef with std::shared_ptr<NodeProperties> in the kernel creation code paths and try to avoid as many copies of NodeDefs as possible. This will in most cases allow sharing the NodeDef between the OpKernel and the graph Node from which it is created.
|
2020-02-18 13:20:06 -08:00 |
|
numeric_op.h
|
Remove changes made to support TFRT-based OpKernel classes in Conv3d kernel.
|
2020-12-09 16:19:36 -08:00 |
|
numeric_types.h
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
op_compatibility_test.cc
|
Add a number of missing headers being transitively pulled in.
|
2020-01-24 15:15:16 -08:00 |
|
op_def_builder_test.cc
|
|
|
|
op_def_builder.cc
|
Update OpDefBuilder with an option to allow attr type to be 'any'.
|
2020-10-12 14:41:10 -07:00 |
|
op_def_builder.h
|
Update OpDefBuilder with an option to allow attr type to be 'any'.
|
2020-10-12 14:41:10 -07:00 |
|
op_def_util_test.cc
|
|
|
|
op_def_util.cc
|
tensorflow/core resolutions set 1
|
2020-07-26 23:49:59 +00:00 |
|
op_def_util.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
op_def.proto
|
Set handle data of function inputs and outputs.
|
2020-12-07 20:55:03 -08:00 |
|
op_gen_lib_test.cc
|
|
|
|
op_gen_lib.cc
|
|
|
|
op_gen_lib.h
|
|
|
|
op_kernel_test.cc
|
Internal tests cleanup.
|
2020-10-29 23:45:48 -07:00 |
|
op_kernel.cc
|
Simplify (compile-time) selective registration
|
2020-09-15 14:54:25 -07:00 |
|
op_kernel.h
|
Propagate eager mode stack traces into OpKernelCtx
|
2021-01-20 12:51:53 -08:00 |
|
op_registration_test.cc
|
|
|
|
op_requires.h
|
Add error payload in status.
|
2021-01-07 00:09:29 -08:00 |
|
op_segment_test.cc
|
|
|
|
op_segment.cc
|
This CL optimizes C++11 range-based for loops where the variable is copied in each iteration but it would suffice to obtain it by const reference. This is only applied to loop variables of types that are expensive to copy which means they are not trivially copyable or have a non-trivial copy constructor or destructor.
|
2020-04-07 14:38:02 -07:00 |
|
op_segment.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
op.cc
|
Simplify (compile-time) selective registration
|
2020-09-15 14:54:25 -07:00 |
|
op.h
|
Simplify (compile-time) selective registration
|
2020-09-15 14:54:25 -07:00 |
|
ops_util.cc
|
|
|
|
ops_util.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
partial_tensor_shape_test.cc
|
Add FooWithStatus methods for all Foo methods with CHECK fails.
|
2021-01-05 14:31:06 -08:00 |
|
partial_tensor_shape.h
|
|
|
|
queue_interface.h
|
|
|
|
reader_base.cc
|
Removed USE_TSTRING usage.
|
2020-01-28 22:26:26 -08:00 |
|
reader_base.h
|
|
|
|
reader_base.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
reader_interface.h
|
|
|
|
reader_op_kernel.h
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
register_types_traits.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
register_types.h
|
Improve GPU dtype coverage of concat
|
2020-09-24 00:52:50 -07:00 |
|
registration_options.h.tpl
|
Add a Bazel-level config setting for experimental changes to registration
|
2020-10-20 11:33:35 -07:00 |
|
remote_fused_graph_execute_info.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
rendezvous_test.cc
|
Merge pull request #44408 from Molkree:typos_in_core_dir
|
2020-11-06 15:01:58 -08:00 |
|
rendezvous.cc
|
Add Ref/Unref to make sure rendezvous outlives its cancellation callback.
|
2020-10-02 11:23:42 -07:00 |
|
rendezvous.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
resource_handle.cc
|
[Cleanup] Remove allowed_devices of ResourceHandle since it's no longer used.
|
2020-06-22 13:18:43 -07:00 |
|
resource_handle.h
|
Propagate the frame where the resource var was created to resource var handles
|
2021-01-25 14:25:35 -08:00 |
|
resource_handle.proto
|
[Cleanup] Remove allowed_devices of ResourceHandle since it's no longer used.
|
2020-06-22 13:18:43 -07:00 |
|
resource_mgr_test.cc
|
[Cleanup] Remove allowed_devices of ResourceHandle since it's no longer used.
|
2020-06-22 13:18:43 -07:00 |
|
resource_mgr.cc
|
Propagate the frame where the resource var was created to resource var handles
|
2021-01-25 14:25:35 -08:00 |
|
resource_mgr.h
|
Propagate the frame where the resource var was created to resource var handles
|
2021-01-25 14:25:35 -08:00 |
|
resource_op_kernel_test.cc
|
Prefixing TensorFlow thread annotation macros with TF_.
|
2020-03-05 08:42:01 -08:00 |
|
resource_op_kernel.h
|
Move uses of tensorflow::MakeTypeIndex() to tensorflow::TypeIndex::Make.
|
2020-06-23 12:50:42 -07:00 |
|
resource_var.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
rng_alg.h
|
Changes StatelessRandomGetKeyCounterAlg to pick RNG_DEFAULT on TPU instead of RNG_THREE_FRY. The latter causes an OOM for StatelessRandomNormalV2 because of excessive TPU padding. The old V1 ops also pick RNG_DEFAULT.
|
2020-11-06 13:52:26 -08:00 |
|
run_handler_test.cc
|
Refactor code for run handler thread pool, and add some unit tests.
|
2020-03-13 15:16:18 -07:00 |
|
run_handler_util_test.cc
|
Implement an alternative setenv/unsetenv, as it is not available on windows.
|
2020-01-08 11:27:52 -08:00 |
|
run_handler_util.cc
|
[RunHandler] Minor optimizations to environment variable handling.
|
2020-04-28 16:07:48 -07:00 |
|
run_handler_util.h
|
[RunHandler] Minor optimizations to environment variable handling.
|
2020-04-28 16:07:48 -07:00 |
|
run_handler.cc
|
[RunHandler] Minor optimizations to environment variable handling.
|
2020-04-28 16:07:48 -07:00 |
|
run_handler.h
|
Refactor code for run handler thread pool, and add some unit tests.
|
2020-03-13 15:16:18 -07:00 |
|
selective_registration_test.cc
|
Disable InitOnStartupTest.ConditionalImmediate and InitOnStartupTest.Conditional which fail on windows.
|
2020-09-24 12:27:37 -07:00 |
|
selective_registration.h
|
Add a Bazel-level config setting for experimental changes to registration
|
2020-10-20 11:33:35 -07:00 |
|
session_state.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
shape_inference_test.cc
|
Some const declarations changed to constexpr
|
2020-04-21 10:21:10 -07:00 |
|
shape_inference_testutil_test.cc
|
|
|
|
shape_inference_testutil.cc
|
|
|
|
shape_inference_testutil.h
|
Fix for adhering to latest clang style guide.
|
2020-05-12 14:08:36 -07:00 |
|
shape_inference.cc
|
Add shape inference rule for XlaPad.
|
2020-12-30 14:15:33 -08:00 |
|
shape_inference.h
|
Add shape inference rule for XlaPad.
|
2020-12-30 14:15:33 -08:00 |
|
shared_ptr_variant.h
|
|
|
|
stats_aggregator.h
|
minor spelling tweaks
|
2020-02-27 15:42:16 +09:00 |
|
step_stats.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
summary.proto
|
Restrict scalar data class to DT_FLOAT tensors
|
2021-01-22 11:29:49 -08:00 |
|
tensor_description.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
tensor_key.h
|
Pre-allocate memory for vectors where size is known.
|
2021-02-10 20:07:07 -08:00 |
|
tensor_reference.h
|
Remove unused unique_tensor_references library.
|
2020-03-12 16:35:29 -07:00 |
|
tensor_shape_test.cc
|
Add FooWithStatus methods for all Foo methods with CHECK fails.
|
2021-01-05 14:31:06 -08:00 |
|
tensor_shape.cc
|
Add FooWithStatus methods for all Foo methods with CHECK fails.
|
2021-01-05 14:31:06 -08:00 |
|
tensor_shape.h
|
Add FooWithStatus methods for all Foo methods with CHECK fails.
|
2021-01-05 14:31:06 -08:00 |
|
tensor_shape.proto
|
Merge pull request #36387 from jhseu:go_fix
|
2020-02-03 17:45:49 -08:00 |
|
tensor_slice_test.cc
|
|
|
|
tensor_slice.cc
|
|
|
|
tensor_slice.h
|
|
|
|
tensor_slice.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
tensor_test.cc
|
Internal tests cleanup.
|
2020-10-29 23:45:48 -07:00 |
|
tensor_testutil_test.cc
|
Make NaNs equal for testing
|
2021-02-09 03:19:22 -08:00 |
|
tensor_testutil.cc
|
Qualify calls to isnan
|
2021-02-10 09:16:32 -08:00 |
|
tensor_testutil.h
|
fix typos in core directory
|
2020-10-29 02:52:55 +03:00 |
|
tensor_types.h
|
|
|
|
tensor_util_test.cc
|
|
|
|
tensor_util.cc
|
More descriptive error message for Concat.
|
2020-06-12 10:15:05 -07:00 |
|
tensor_util.h
|
Clean up OpKernel 1.
|
2020-01-29 15:49:57 -08:00 |
|
tensor.cc
|
[tensorflow] Add a Tensor constructor that allows to transfer TensorBuffer ownership
|
2021-02-11 09:42:22 -08:00 |
|
tensor.h
|
[tensorflow] Add a Tensor constructor that allows to transfer TensorBuffer ownership
|
2021-02-11 09:42:22 -08:00 |
|
tensor.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
thread_factory.h
|
|
|
|
tracking_allocator_test.cc
|
|
|
|
tracking_allocator.cc
|
|
|
|
tracking_allocator.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
type_index.h
|
Fix issues with TypeIndex on Windows, i.e. hash on the type name where available since this otherwise causes problems when loading different shared libraries with RTLD_LOCAL.
|
2021-01-22 11:22:37 -08:00 |
|
type_traits.h
|
Some const declarations changed to constexpr
|
2020-04-21 10:21:10 -07:00 |
|
typed_allocator.cc
|
|
|
|
typed_allocator.h
|
|
|
|
types_test.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
types.cc
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
types.h
|
Delete SYCL support
|
2020-09-15 11:21:47 -07:00 |
|
types.proto
|
Add pfor converters and basic type information to optionals
|
2020-11-19 09:25:24 -08:00 |
|
variable.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
|
variant_encode_decode.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
variant_op_copy_test.cc
|
|
|
|
variant_op_registry_test.cc
|
Move uses of tensorflow::MakeTypeIndex() to tensorflow::TypeIndex::Make.
|
2020-06-23 12:50:42 -07:00 |
|
variant_op_registry.cc
|
Make "missing variant function" error status a bit more readable
|
2021-02-15 23:48:41 -08:00 |
|
variant_op_registry.h
|
Make "missing variant function" error status a bit more readable
|
2021-02-15 23:48:41 -08:00 |
|
variant_tensor_data.cc
|
Move VariantTensorData::add_tensor template definition from .cc to .h file.
|
2020-11-18 16:38:26 -08:00 |
|
variant_tensor_data.h
|
Move VariantTensorData::add_tensor template definition from .cc to .h file.
|
2020-11-18 16:38:26 -08:00 |
|
variant_test.cc
|
Move uses of tensorflow::MakeTypeIndex() to tensorflow::TypeIndex::Make.
|
2020-06-23 12:50:42 -07:00 |
|
variant.cc
|
|
|
|
variant.h
|
Qualify uses of std::string
|
2020-07-31 12:01:59 -07:00 |
|
versions.cc
|
|
|
|
versions.h
|
|
|
|
versions.proto
|
Give unique go package paths in core/framework to avoid circular dependencies.
|
2020-02-05 13:14:29 -08:00 |
