- 15 6月, 2018 21 次提交
-
-
由 qingqing01 提交于
* Update some doc about layers' API. * Fix format. * Fix example bug in random_data_generator. * Fix example bug in dropout. * Follow comments and some small fix for some examples.
-
由 chengduo 提交于
skip all tests in tests_parallel_executor_crf
-
由 Tao Luo 提交于
Fix dist ut
-
由 Tao Luo 提交于
Polish inline math and duplicable/optional in auto generated doc
-
由 fengjiayi 提交于
-
由 yuyang18 提交于
-
由 tensor-tang 提交于
Doc update
-
由 Yan Chunwei 提交于
-
由 QI JUN 提交于
* fix mac build error * enhance memory optimize transpiler to let users set to some skip opt set of variables
-
由 Yu Yang 提交于
Polish documentation
-
由 yuyang18 提交于
-
由 Shan Yi 提交于
fix deadlink in contribute_to_paddle_cn.rst
-
由 Yan Chunwei 提交于
-
由 chengduo 提交于
Enable more type for SplitOp and ConcatOp
-
由 Xin Pan 提交于
Add API docs.
-
由 chengduo 提交于
Add conv3d/pool3d/conv3d_trans Python API
-
由 Kexin Zhao 提交于
-
由 tensor-tang 提交于
-
由 tensor-tang 提交于
-
由 tensor-tang 提交于
-
由 tensor-tang 提交于
-
- 14 6月, 2018 19 次提交
-
-
由 Tao Luo 提交于
Correct the use of the lstm layer
-
由 gongweibao 提交于
-
由 Yan Chunwei 提交于
-
由 Tao Luo 提交于
initial with only 1 mkl/openblas threads for each pthreads
-
-
由 chengduoZH 提交于
-
-
由 yi.wu 提交于
-
由 yuyang18 提交于
-
由 tensor-tang 提交于
-
由 Tao Luo 提交于
fix compiler error in contrib/tape
-
由 yuyang18 提交于
-
由 Luo Tao 提交于
-
由 Luo Tao 提交于
-
由 tensor-tang 提交于
-
由 tensor-tang 提交于
Add an interface to set the number of threads for math function, and set the default value to 1 for inference.
-
由 fengjiayi 提交于
fix a small error
-
由 Qiyang Min 提交于
* 1. Create buddy allocator in each places before NcclBcast the variables 2. Check the memory usage of ALL gpus rather than the first one * 1. Make NCCLGroupGuard guards only the ncclBcast part, which avoid ncclGroupEnd blocking the exception throwing 2. NOTE the usage of NCCLGroupGuard * Remove the memory usage check of gpus * Fix code style
-
由 Qiao Longfei 提交于
outvar must be create in local scope for prefetch
-