chengduo a6a3b2fbbc
[Speed]Refine ParallelExecutor ()
6 years ago
..
CMakeLists.txt [Speed]Refine ParallelExecutor () 6 years ago
all_reduce_deps_pass.cc polish codes 6 years ago
all_reduce_deps_pass.h update by comment test=develop 6 years ago
all_reduce_op_handle.cc [Speed]Refine ParallelExecutor () 6 years ago
all_reduce_op_handle.h fix unit test cases 6 years ago
alloc_continuous_space_for_grad_pass.cc Fuse AllReduce () 6 years ago
broadcast_op_handle.cc Profiler refine and add CUDA runtime api tracer () 6 years ago
broadcast_op_handle.h [Speed]Refine ParallelExecutor () 6 years ago
broadcast_op_handle_test.cc add fused broadcast op unit test, test=develop 7 years ago
broadcast_op_handle_test.h fix unit test cases 6 years ago
build_strategy.cc [Speed]Refine ParallelExecutor () 6 years ago
build_strategy.h Fuse AllReduce () 6 years ago
computation_op_handle.cc Remove debug info 6 years ago
computation_op_handle.h fix travis-ci format check 6 years ago
container_cast.h Clean Code 7 years ago
cow_ptr.h Fix MixedVector 7 years ago
cow_ptr_test.cc Revert "Revert "Merge pull request from chengduoZH/refine_lod"" 7 years ago
eager_deletion_op_handle.cc fix travis-ci format check 6 years ago
eager_deletion_op_handle.h polish code 6 years ago
eager_deletion_pass.cc fix travis-ci format check 6 years ago
early_delete_op_handle.h add ir memory optimize. () 6 years ago
exception_holder.h refactor(memory): rewrite memory allocation and make it extentable 7 years ago
execution_strategy.h revert test=develop () 6 years ago
fast_threaded_ssa_graph_executor.cc Unified ParallelExecutor and Compiler () 6 years ago
fast_threaded_ssa_graph_executor.h allow compiler to use graph 6 years ago
fetch_op_handle.cc [Speed]Refine ParallelExecutor () 6 years ago
fetch_op_handle.h [Speed]Refine ParallelExecutor () 6 years ago
fuse_all_reduce_op_pass.cc Fuse AllReduce () 6 years ago
fused_all_reduce_op_handle.cc Add unit test for fuse all reduce () 6 years ago
fused_all_reduce_op_handle.h Fuse AllReduce () 6 years ago
fused_broadcast_op_handle.cc Profiler refine and add CUDA runtime api tracer () 6 years ago
fused_broadcast_op_handle.h fix unit test cases 6 years ago
fused_broadcast_op_handle_test.cc Tests - add some missing to_string calls 6 years ago
gather_op_handle.cc Hide varhandle members. () 6 years ago
gather_op_handle.h op compose node and update nodes. 7 years ago
gather_op_handle_test.cc fix some tests. 6 years ago
graph_test_base.h Polish code style 6 years ago
inplace_op_pass.cc Add some fixme. test=develop 6 years ago
inplace_op_pass.h add details. test=develop 6 years ago
memory_optimize_helper.cc 1. disable reuse SELECTED_ROWS type variable () 6 years ago
memory_optimize_helper.h fix default value. test=develop 6 years ago
memory_optimize_helper_test.cc allow compiler to use graph 6 years ago
memory_optimize_pass.cc Add some fixme. test=develop 6 years ago
memory_optimize_pass.h add reference. test=develop 6 years ago
modify_op_lock_and_record_event_pass.cc Revert the changes of VLOG 6 years ago
modify_op_lock_and_record_event_pass.h remove_lock_in_some_ops 7 years ago
multi_devices_graph_check_pass.cc Refactor MultiDevSSAGraphBuilder () 6 years ago
multi_devices_graph_pass.cc [Speed]Refine ParallelExecutor () 6 years ago
multi_devices_graph_pass.h Fuse AllReduce () 6 years ago
multi_devices_graph_print_pass.cc Hide varhandle members. () 6 years ago
multi_devices_graph_print_pass.h delete graph print pass. test=develop 6 years ago
multi_devices_helper.cc code clean up and renaming 7 years ago
multi_devices_helper.h Fuse AllReduce () 6 years ago
op_graph_view.cc polish code 6 years ago
op_graph_view.h fix bug 6 years ago
op_handle_base.cc [Speed]Refine ParallelExecutor () 6 years ago
op_handle_base.h refine pg execution 6 years ago
op_registry.h Polish code style 6 years ago
parallel_ssa_graph_executor.cc add parallel graph dist test () 6 years ago
parallel_ssa_graph_executor.h polish 6 years ago
reduce_and_gather.h Fuse AllReduce () 6 years ago
reduce_op_handle.cc Profiler refine and add CUDA runtime api tracer () 6 years ago
reduce_op_handle.h Add reduce sparse tensor feature. () 6 years ago
reduce_op_handle_test.cc fix unit test cases 6 years ago
reference_count_pass.cc fix travis-ci format check 6 years ago
reference_count_pass.h refactor eager deletion 6 years ago
reference_count_pass_helper.cc partial gc 1st version 6 years ago
reference_count_pass_helper.h fix travis-ci format check 6 years ago
rpc_op_handle.cc Hide varhandle members. () 6 years ago
rpc_op_handle.h op compose node and update nodes. 7 years ago
scale_loss_grad_op_handle.cc Hide varhandle members. () 6 years ago
scale_loss_grad_op_handle.h Fp16 training () 6 years ago
scope_buffered_ssa_graph_executor.cc Profiler refine and add CUDA runtime api tracer () 6 years ago
scope_buffered_ssa_graph_executor.h Fix wait stream two times bug 6 years ago
sequential_execution_pass.cc polish codes 6 years ago
sequential_execution_pass.h add details. test=develop 6 years ago
ssa_graph_executor.cc fix 6 years ago
ssa_graph_executor.h clean1 6 years ago
threaded_ssa_graph_executor.cc [Speed]Refine ParallelExecutor () 6 years ago
threaded_ssa_graph_executor.h [Speed]Refine ParallelExecutor () 6 years ago
var_handle.cc code cleanup test=develop 6 years ago
var_handle.h [Speed]Refine ParallelExecutor () 6 years ago
variable_visitor.cc rewrite variable type 6 years ago
variable_visitor.h follow comments and clean code 7 years ago
while_op_eager_deletion_pass.cc enhance gc 6 years ago