“80c68d38ff3c59e12f48b4e4e88c24c89568fc0a”上不存在“paddle/legacy/cuda/include/hl_top_k.h”
- 11 1月, 2021 1 次提交
-
-
由 WangXi 提交于
* Optimization grad merge performance (#29784) * [fleet] combine amp and gradient merge, test=develop (#30086) * fix assign_op_xpu concat_op_xpu warining (#30120) Co-authored-by: Nliuyuhui <liuyuhui@baidu.com>
-
- 13 10月, 2020 1 次提交
-
-
由 mapingshuo 提交于
* support gradient merge with recompute, test=develop test=develop
-
- 10 9月, 2020 1 次提交
-
-
由 123malin 提交于
* parameter_server_optimizer support auto_strategy
-
- 07 9月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* add auto parallel L1 implementation test=develop
-
- 29 8月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* fix api document
-
- 21 8月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* consider the combination of different strategies to work together
-
- 17 8月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* add check approval test=develop
-
- 13 8月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* move paddle.fleet to paddle.distributed.fleet
-
- 30 7月, 2020 1 次提交
-
-
由 mapingshuo 提交于
* add gradient Merge optimizer to meta, test=develop
-
- 29 7月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
* refine strategy compiler and meta optimizers make async as a_sync
-
- 20 7月, 2020 1 次提交
-
-
由 Dong Daxiang 提交于
refactor fleet api under paddle.fleet update DistributedStrategy
-