| .. |
|
details
|
Open fuse optimization ops (#18741)
|
6 years ago |
|
fleet
|
add clear_model interface in fleetwrapper (#18815)
|
6 years ago |
|
io
|
fix runtime error
|
6 years ago |
|
ir
|
Open fuse optimization ops (#18741)
|
6 years ago |
|
.gitignore
|
Windows/hidden file (#13139)
|
7 years ago |
|
CMakeLists.txt
|
Support memory eager deletion on recurrent OP (#17710)
|
6 years ago |
|
archive.h
|
dataset (#17973)
|
6 years ago |
|
array.h
|
fix py35 compile error
|
6 years ago |
|
async_executor.cc
|
Pipeline Concurrency (#17402)
|
6 years ago |
|
async_executor.h
|
fix dataset float32 type problem
|
6 years ago |
|
attribute.cc
|
shape type to int64_t, test=develop
|
6 years ago |
|
attribute.h
|
fix #17430: int64类型的attr训练非预期 (#18264)
|
6 years ago |
|
block_desc.cc
|
Add missing headers
|
6 years ago |
|
block_desc.h
|
invoke backward_hooks after reduce op's depcounts map
|
6 years ago |
|
blocking_queue.h
|
add save/load model, shrink table, cvm, config file & fix pull dense bug (#17118)
|
6 years ago |
|
channel.h
|
dataset (#17973)
|
6 years ago |
|
commit.h.in
|
add version support (#15469)
|
6 years ago |
|
data_device_transform.cc
|
Revert the changes of VLOG
|
6 years ago |
|
data_device_transform.h
|
…
|
|
|
data_device_transform_test.cu
|
rewrite variable type
|
6 years ago |
|
data_feed.cc
|
support patch data, add load_one_table, fix bug (#18509)
|
6 years ago |
|
data_feed.h
|
support patch data, add load_one_table, fix bug (#18509)
|
6 years ago |
|
data_feed.proto
|
add gpu training for Executor.train_from_dataset
|
6 years ago |
|
data_feed_factory.cc
|
Pipeline Concurrency (#17402)
|
6 years ago |
|
data_feed_factory.h
|
AsyncExecutor (#14627)
|
6 years ago |
|
data_feed_test.cc
|
fix async_executor problem and remove some unnecessary testcase, fix trainer_desc import problem
|
6 years ago |
|
data_layout.h
|
Mkldnn layout (#11040)
|
7 years ago |
|
data_layout_transform.cc
|
enable mkldnn primitive reuse for platform reorder (#17826)
|
6 years ago |
|
data_layout_transform.h
|
Fix/gcc 4.8 ubt link error (#18558)
|
6 years ago |
|
data_layout_transform_test.cc
|
…
|
|
|
data_set.cc
|
support patch data, add load_one_table, fix bug (#18509)
|
6 years ago |
|
data_set.h
|
support patch data, add load_one_table, fix bug (#18509)
|
6 years ago |
|
data_transform.cc
|
[MKL-DNN] Tensor modifications revert (#16462)
|
6 years ago |
|
data_transform.h
|
Rename TransferData -> TransformData
|
7 years ago |
|
data_type.cc
|
Change tensor uses proto::VarType::type
|
6 years ago |
|
data_type.h
|
Fix/gcc 4.8 ubt link error (#18558)
|
6 years ago |
|
data_type_test.cc
|
Fix ut
|
6 years ago |
|
data_type_transform.cc
|
Feature/template (#13093)
|
7 years ago |
|
data_type_transform.cu
|
revert
|
6 years ago |
|
data_type_transform.h
|
…
|
|
|
data_type_transform_test.cc
|
…
|
|
|
data_type_transform_test.cu
|
…
|
|
|
dataset_factory.cc
|
dataset (#17973)
|
6 years ago |
|
dataset_factory.h
|
dataset (#17973)
|
6 years ago |
|
ddim.cc
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
ddim.h
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
ddim_test.cc
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
device_worker.cc
|
dataset (#17973)
|
6 years ago |
|
device_worker.h
|
Fix shrink-dense and add scale-datanorm (#18746)
|
6 years ago |
|
device_worker_factory.cc
|
Pipeline Concurrency (#17402)
|
6 years ago |
|
device_worker_factory.h
|
fix class register problem
|
6 years ago |
|
device_worker_test.cc
|
fix dataset float32 type problem
|
6 years ago |
|
dim.h
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
dim_test.cu
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
dist_multi_trainer.cc
|
dataset (#17973)
|
6 years ago |
|
dlpack_tensor.cc
|
merge develop
|
6 years ago |
|
dlpack_tensor.h
|
rewrite ddim
|
6 years ago |
|
dlpack_tensor_test.cc
|
Change tensor uses proto::VarType::type
|
6 years ago |
|
downpour_worker.cc
|
Fix shrink-dense and add scale-datanorm (#18746)
|
6 years ago |
|
eigen.h
|
Windows/online (#14474)
|
6 years ago |
|
eigen_test.cc
|
…
|
|
|
executor.cc
|
Support memory eager deletion on recurrent OP (#17710)
|
6 years ago |
|
executor.h
|
polish_executor_and_add_ctx_cache (#17536)
|
6 years ago |
|
executor_gc_helper.cc
|
fix env variable settting bug
|
6 years ago |
|
executor_gc_helper.h
|
try to fix ci error
|
6 years ago |
|
executor_thread_worker.cc
|
fix code style & add dataset testcase
|
6 years ago |
|
executor_thread_worker.h
|
add TrainFilesWithTimer in async_executor
|
6 years ago |
|
expect.h
|
dataset (#17973)
|
6 years ago |
|
feed_fetch_method.cc
|
fix potential bug (#15688)
|
6 years ago |
|
feed_fetch_method.h
|
Revert "Merge pull request #14798 from PaddlePaddle/revert-14786-revert-14782-revert-14398-imperative"
|
6 years ago |
|
feed_fetch_type.h
|
…
|
|
|
framework.proto
|
polish
|
6 years ago |
|
garbage_collector.cc
|
fix env variable settting bug
|
6 years ago |
|
garbage_collector.h
|
Refine model gpu memory (#16993)
|
6 years ago |
|
grad_op_desc_maker.h
|
check default grad maker
|
6 years ago |
|
hogwild_worker.cc
|
support patch data, add load_one_table, fix bug (#18509)
|
6 years ago |
|
inlined_vector.h
|
Fix allocator bug (#16712)
|
6 years ago |
|
inlined_vector_test.cc
|
Fix allocator bug (#16712)
|
6 years ago |
|
inplace_op_inference.h
|
Feature/buffer_shared_inplace (#17911)
|
6 years ago |
|
inplace_op_inference_test.cc
|
Feature/mem opt pass refactor (#18735)
|
6 years ago |
|
library_type.h
|
…
|
|
|
lod_rank_table.cc
|
Revert "Optimize while_op when is_test is true. (#15811)" (#15968)
|
6 years ago |
|
lod_rank_table.h
|
…
|
|
|
lod_tensor.cc
|
add __str__ method for tensor and lodtensor to support print test=dev… (#17588)
|
6 years ago |
|
lod_tensor.h
|
Support seq len equal to 0 in sequence ops (#16935)
|
6 years ago |
|
lod_tensor_array.h
|
fix tensor array bug (#14166)
|
6 years ago |
|
lod_tensor_test.cc
|
Code clean of Allocator (#17602)
|
6 years ago |
|
lod_tensor_test.cu
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into complete_py_reader_python
|
7 years ago |
|
mixed_vector.h
|
Feature/auto_growth_allocator (#18561)
|
6 years ago |
|
mixed_vector_test.cc
|
Revert the changes of VLOG
|
6 years ago |
|
mixed_vector_test.cu
|
fix mixed_vector_test
|
7 years ago |
|
multi_trainer.cc
|
dataset (#17973)
|
6 years ago |
|
naive_executor.cc
|
fea/infer memory optim2 (#14953)
|
6 years ago |
|
naive_executor.h
|
Combine Inference Analysis with IR (#13914)
|
6 years ago |
|
naive_executor_test.cc
|
Combine Inference Analysis with IR (#13914)
|
6 years ago |
|
no_need_buffer_vars_inference.h
|
add op registry type
|
6 years ago |
|
op_desc.cc
|
Security issue (#16774)
|
6 years ago |
|
op_desc.h
|
Add DGC(Deep Gradient Compression) interface. (#15841)
|
6 years ago |
|
op_info.cc
|
check default grad maker
|
6 years ago |
|
op_info.h
|
check default grad maker
|
6 years ago |
|
op_kernel_type.cc
|
allow customize kernel selection
|
6 years ago |
|
op_kernel_type.h
|
allow customize kernel selection
|
6 years ago |
|
op_kernel_type_test.cc
|
Fix ut
|
6 years ago |
|
op_proto_maker.cc
|
supports collective training with programs (#18392)
|
6 years ago |
|
op_proto_maker.h
|
supports collective training with programs (#18392)
|
6 years ago |
|
op_proto_maker_test.cc
|
clean up after the changes have been stopped for so long.
|
6 years ago |
|
op_registry.cc
|
code style fix, test=develop (#15045)
|
6 years ago |
|
op_registry.h
|
Enable function coverage for U8/S8 ConvMKLDNNOpKernel
|
6 years ago |
|
op_registry_test.cc
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into complete_py_reader_python
|
7 years ago |
|
operator.cc
|
not use transferscope cache in cpu case (#18578)
|
6 years ago |
|
operator.h
|
not use transferscope cache in cpu case (#18578)
|
6 years ago |
|
operator_kernel_configs.h
|
Remove attribute in Allocator::Allocate (#17878)
|
6 years ago |
|
operator_test.cc
|
Polish code
|
6 years ago |
|
parallel_executor.cc
|
Feature/mem opt pass refactor (#18735)
|
6 years ago |
|
parallel_executor.h
|
Feature/buffer_shared_inplace (#17911)
|
6 years ago |
|
pipeline_trainer.cc
|
dataset (#17973)
|
6 years ago |
|
program_desc.cc
|
fix some compiler warning
|
6 years ago |
|
program_desc.h
|
clean
|
7 years ago |
|
program_desc_test.cc
|
fix typo and warning in analyzer_resnet50_test
|
6 years ago |
|
proto_desc.h
|
…
|
|
|
prune.cc
|
clean unused inference_optimize c++ implementation
|
7 years ago |
|
prune.h
|
clean unused inference_optimize c++ implementation
|
7 years ago |
|
prune_test.cc
|
…
|
|
|
pull_dense_worker.cc
|
add save/load model, shrink table, cvm, config file & fix pull dense bug (#17118)
|
6 years ago |
|
python_headers.h
|
invoke backward_hooks after reduce op's depcounts map
|
6 years ago |
|
reader.cc
|
pure async mode train
|
6 years ago |
|
reader.h
|
code format test=develop
|
6 years ago |
|
reader_test.cc
|
fix typo and warning in analyzer_resnet50_test
|
6 years ago |
|
rw_lock.h
|
Remove unnessesary code
|
6 years ago |
|
rw_lock_test.cc
|
add rw lock test (#12752)
|
7 years ago |
|
scope.cc
|
Feature/buffer_shared_inplace (#17911)
|
6 years ago |
|
scope.h
|
Feature/buffer_shared_inplace (#17911)
|
6 years ago |
|
scope_pool.cc
|
add scope_pool
|
6 years ago |
|
scope_pool.h
|
add scope_pool
|
6 years ago |
|
scope_test.cc
|
…
|
|
|
section_worker.cc
|
Pipeline Concurrency (#17402)
|
6 years ago |
|
selected_rows.cc
|
Change tensor uses proto::VarType::type
|
6 years ago |
|
selected_rows.h
|
add more files to protected file list
|
6 years ago |
|
selected_rows_test.cc
|
add is_test for lookup_sparse_table
|
6 years ago |
|
shape_inference.cc
|
move more and fix while
|
6 years ago |
|
shape_inference.h
|
Merge pull request #14836 from sneaxiy/feature/py_func
|
6 years ago |
|
tensor.cc
|
Remove attribute in Allocator::Allocate (#17878)
|
6 years ago |
|
tensor.h
|
Feature/buffer_shared_inplace (#17911)
|
6 years ago |
|
tensor_impl.h
|
Fix/gcc 4.8 ubt link error (#18558)
|
6 years ago |
|
tensor_test.cc
|
Feature/auto_growth_allocator (#18561)
|
6 years ago |
|
tensor_util.cc
|
add __str__ method for tensor and lodtensor to support print test=dev… (#17588)
|
6 years ago |
|
tensor_util.cu
|
revert
|
6 years ago |
|
tensor_util.h
|
add __str__ method for tensor and lodtensor to support print test=dev… (#17588)
|
6 years ago |
|
tensor_util_test.cc
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rewrite_allocation
|
6 years ago |
|
tensor_util_test.cu
|
enhance isinf/isnan in tensor util, avoid copy back to cpu (#12688)
|
7 years ago |
|
threadpool.cc
|
clean code
|
6 years ago |
|
threadpool.h
|
update by comment test=develop
|
6 years ago |
|
threadpool_test.cc
|
clean code
|
6 years ago |
|
trainer.cc
|
refine device_worker and trainer code
|
6 years ago |
|
trainer.h
|
dataset (#17973)
|
6 years ago |
|
trainer_desc.proto
|
Fix shrink-dense and add scale-datanorm (#18746)
|
6 years ago |
|
trainer_factory.cc
|
Pipeline Concurrency (#17402)
|
6 years ago |
|
trainer_factory.h
|
fix class register problem
|
6 years ago |
|
trainer_test.cc
|
fix dataset float32 type problem
|
6 years ago |
|
transfer_scope_cache.cc
|
use static variable to do cache instead of thread local in thread frequent switching case (#18428)
|
6 years ago |
|
transfer_scope_cache.h
|
add transfer_scope_cache unit-test (#18467)
|
6 years ago |
|
tuple.h
|
clean up channel
|
7 years ago |
|
tuple_test.cc
|
…
|
|
|
type_defs.h
|
Add use_cuda to inplace pass (#17205)
|
6 years ago |
|
unroll_array_ops.h
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
unroll_array_ops_test.cc
|
Clean unused code of dim and place (#18565)
|
6 years ago |
|
var_desc.cc
|
update alloc_continuous_space_for_grad_pass (#18287)
|
6 years ago |
|
var_desc.h
|
update alloc_continuous_space_for_grad_pass (#18287)
|
6 years ago |
|
var_type.h
|
polish code
|
6 years ago |
|
var_type_inference.h
|
Correct the lod level of compiled time in lod_reset (#16790)
|
6 years ago |
|
var_type_inference_test.cc
|
Polish code style
|
6 years ago |
|
var_type_traits.cc
|
Fix reinitialized ncclid error! (#18025)
|
6 years ago |
|
var_type_traits.h
|
Fix reinitialized ncclid error! (#18025)
|
6 years ago |
|
var_type_traits_test.cc
|
Fix reinitialized ncclid error! (#18025)
|
6 years ago |
|
variable.h
|
polish code
|
6 years ago |
|
variable_helper.cc
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
|
6 years ago |
|
variable_helper.h
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
|
6 years ago |
|
variable_test.cc
|
rewrite variable type
|
6 years ago |
|
version.cc
|
clean
|
7 years ago |
|
version.h
|
clean
|
7 years ago |
|
version_test.cc
|
refine
|
7 years ago |
Thunderbrook