16596f6498
Fix Paddle Cloud role maker ( #20860 )
...
* fix PaddleCloud Role maker & add warning in distribute transpiler & change rpc_retry_times
6 years ago
940c6ff1c8
Fix communicator slow bug & fix communicator stop bug ( #20366 )
...
* test=develop,Fix communicator slow bug
* test=develop, delete if() in stop_worker()
* test=develop
* fix UT, test=develop
* fix bug in fetch handler, test=develop
* fix bug in fetch handler, test=develop
* test=develop, fix fetch barrier bug
* test=develop, bug fix
* test=develop, bug fix
* test=develop, fix bug
6 years ago
728ec1b43d
Add GEO-SGD distribute training algorithm ( #20018 )
...
* refector geo sgd & communicator
6 years ago
a25a716e87
Optimize fleet API: add input check for some interfaces ( #18971 )
...
* fleet api add input check, test=develop
6 years ago
65c7368400
Fix the correctness of async mode at distributed training ( #18863 )
...
* fix correctness of the communicator
* fix a bug in send thread when sending var context is empty, test=develop
* add lookup_table_prefetch_op and prefetch optimize, test=develop
* remove remote prefetch GPU supported
* word2vec force with CPU, test=develop
* test dist remote lookup table force with CPU, test=develop
6 years ago
86f0591175
Remove node_num function. ( #19167 )
...
node_num is not needed for users, so remove them and fix the bugs about it!
6 years ago
29d8781240
Polish fleet API to support cuda collective mode and nccl2 mode. ( #18966 )
...
Polish fleet API to support cuda collective mode and nccl2 mode
6 years ago
357311fdb7
make fleet support mpi job submit directly ( #18441 )
...
make fleet support mpi job submit directly.
6 years ago
999d9a59a5
fix communicator with pyreader ( #18350 )
...
* add is_runnning in communicator, test=develop
6 years ago
4c735f24ea
fix bug in fleet, test=develop ( #18058 )
6 years ago
101f74cb19
fix save/load in fleet ( #17675 )
...
* fix save/load in Fleet
* add UT framework of Fleet
6 years ago
Chengmo
123malin
tangwei12
gongweibao
guru4elephant