| .. |
|
allreduce_fusion
|
!586 enable/disable allreduce_fusion
|
2020-04-24 16:02:53 +08:00 |
|
auto_parallel
|
use DeviceMemory for memory control
|
2020-04-23 17:55:17 +08:00 |
|
graph_util
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
ops_info
|
use DeviceMemory for memory control
|
2020-04-23 17:55:17 +08:00 |
|
strategy_checkpoint
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
tensor_layout
|
fix the codex and bot warnings
|
2020-04-23 19:14:38 +08:00 |
|
context.cc
|
enable/disable allreduce_fusion
|
2020-04-24 10:34:59 +08:00 |
|
context.h
|
enable/disable allreduce_fusion
|
2020-04-24 10:34:59 +08:00 |
|
costmodel_context.cc
|
Change 'NOT_FULLY_USE_DEVICES' to 'FULLY_USE_DEVICES' and make ALL-1 user-specified-strategy valid in auto-parallel
|
2020-04-15 10:14:46 +08:00 |
|
costmodel_context.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
device.h
|
use std::vector instead of std::list to promote performance for parallel module
|
2020-03-31 15:16:00 +08:00 |
|
device_manager.cc
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
device_manager.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
device_matrix.cc
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
device_matrix.h
|
fix the codex and bot warnings
|
2020-04-23 19:14:38 +08:00 |
|
dynamic_creator.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
group_manager.cc
|
sort include file in parallel dir
|
2020-04-02 15:14:47 +08:00 |
|
group_manager.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
node_check.cc
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
node_check.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
|
status.h
|
fix_coding_style_check_warning
|
2020-04-15 16:15:45 +08:00 |
|
step_auto_parallel.cc
|
use DeviceMemory for memory control
|
2020-04-23 17:55:17 +08:00 |
|
step_auto_parallel.h
|
fix_coding_style_check_warning
|
2020-04-15 16:15:45 +08:00 |
|
step_parallel.cc
|
!264 static_analysis: remove useless cache in TrivialPrimEvaluator and add cache for PythonPrimEvaluator
|
2020-04-24 09:22:33 +08:00 |
|
step_parallel.h
|
support multiple subgraphs
|
2020-04-22 17:40:56 +08:00 |
|
strategy.h
|
update clang format rule
|
2020-04-21 21:21:19 +08:00 |
00859ae119