You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Zeng Jinle a9c8bdad7b refine pe codes, test=develop (#20479)		6 years ago
..
CMakeLists.txt	Add place deps for fused_all_reduce_op_handle (#20077)	6 years ago
all_reduce_op_handle.cc	Enable fused_all_reduce_op_handle support GPU and CPU Gradients (#19418)	6 years ago
all_reduce_op_handle.h	Enable fused_all_reduce_op_handle support GPU and CPU Gradients (#19418)	6 years ago
async_ssa_graph_executor.cc	the integrated communicator (#19849)	6 years ago
async_ssa_graph_executor.h	Feature/buffer_shared_inplace (#17911)	6 years ago
broadcast_op_handle.cc	fix reduce and broadcast to avoid multi-stream, test=develop (#19889)	6 years ago
broadcast_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
broadcast_op_handle_test.cc	…
broadcast_op_handle_test.h	Feature/buffer_shared_inplace (#17911)	6 years ago
build_strategy.cc	Fix warning info of build_strategy (#19805)	6 years ago
build_strategy.h	disable fuse_all_optimizer_ops (#19966)	6 years ago
computation_op_handle.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
computation_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
container_cast.h	…
cow_ptr.h	…
cow_ptr_test.cc	…
dgc_const_values.h	Polish DGC code (#16818)	6 years ago
eager_deletion_op_handle.cc	[Bug fix] Disable memory reuse on feeded variables (#19835)	6 years ago
eager_deletion_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
exception_holder.h	code format test=develop	7 years ago
execution_strategy.h	Increase num_iteration_per_drop_scope (#19075)	6 years ago
fast_threaded_ssa_graph_executor.cc	fix fast pe to run highest priority ops first, test=develop (#19575)	6 years ago
fast_threaded_ssa_graph_executor.h	Feature/buffer_shared_inplace (#17911)	6 years ago
fetch_barrier_op_handle.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
fetch_barrier_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
fetch_op_handle.cc	merge empty lod tensor, test=develop (#19228)	6 years ago
fetch_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
fused_all_reduce_op_handle.cc	Add place deps for fused_all_reduce_op_handle (#20077)	6 years ago
fused_all_reduce_op_handle.h	Enable fused_all_reduce_op_handle support GPU and CPU Gradients (#19418)	6 years ago
fused_broadcast_op_handle.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
fused_broadcast_op_handle.h	fix unit test cases	7 years ago
fused_broadcast_op_handle_test.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
gather_op_handle.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
gather_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
gather_op_handle_test.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
graph_test_base.h	Polish code style	7 years ago
multi_devices_helper.cc	…
multi_devices_helper.h	Enable fused_all_reduce_op_handle support GPU and CPU Gradients (#19418)	6 years ago
nccl_op_handle.h	Polish backwards optimizer dependency codes and use more default values. (#18255)	6 years ago
op_handle_base.cc	refine pe codes, test=develop (#20479)	6 years ago
op_handle_base.h	Feature/mem opt pass refactor (#18735)	6 years ago
op_registry.h	Add use_cuda to inplace pass (#17205)	6 years ago
parallel_ssa_graph_executor.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
parallel_ssa_graph_executor.h	Feature/buffer_shared_inplace (#17911)	6 years ago
reduce_and_gather.h	fix reduce and broadcast to avoid multi-stream, test=develop (#19889)	6 years ago
reduce_op_handle.cc	fix reduce and broadcast to avoid multi-stream, test=develop (#19889)	6 years ago
reduce_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
reduce_op_handle_test.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
rpc_op_handle.cc	Feature/buffer_shared_inplace (#17911)	6 years ago
rpc_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
scale_loss_grad_op_handle.cc	remove -Wmaybe-uninitialized warning (#19653)	6 years ago
scale_loss_grad_op_handle.h	Feature/buffer_shared_inplace (#17911)	6 years ago
scope_buffered_monitor.cc	Delete local execution scopes (#19749)	6 years ago
scope_buffered_monitor.h	Delete local execution scopes (#19749)	6 years ago
scope_buffered_ssa_graph_executor.cc	clean tensor array (#19930)	6 years ago
scope_buffered_ssa_graph_executor.h	clean tensor array (#19930)	6 years ago
share_tensor_buffer_functor.cc	[Bug fix] Disable memory reuse on feeded variables (#19835)	6 years ago
share_tensor_buffer_functor.h	Feature/mem opt pass refactor (#18735)	6 years ago
share_tensor_buffer_op_handle.cc	Feature/mem opt pass refactor (#18735)	6 years ago
share_tensor_buffer_op_handle.h	Feature/mem opt pass refactor (#18735)	6 years ago
sparse_all_reduce_op_handle.cc	fix cuda dev_ctx by event, test=develop (#20553)	6 years ago
sparse_all_reduce_op_handle.h	fix cuda dev_ctx by event, test=develop (#20553)	6 years ago
ssa_graph_executor.cc	[Speed] Refine the Executor when the num_thread=1 (#17405)	6 years ago
ssa_graph_executor.h	[Speed] Refine the Executor when the num_thread=1 (#17405)	6 years ago
threaded_ssa_graph_executor.cc	fix PE fetch bug (#18644)	6 years ago
threaded_ssa_graph_executor.h	Feature/buffer_shared_inplace (#17911)	6 years ago
var_handle.cc	Add DGC(Deep Gradient Compression) interface. (#15841)	6 years ago
var_handle.h	Feature/mem opt pass refactor (#18735)	6 years ago
variable_visitor.cc	rewrite variable type	7 years ago
variable_visitor.h	…