- 28 9月, 2020 1 次提交
-
-
由 root 提交于
update initializer API, including: NormalInitializer, TruncatedNormalInitializer, UniformInitializer, XavierNormalInitializer, XavierUniformInitializer, NumpyArrayInitializer.
-
- 27 9月, 2020 7 次提交
-
-
由 YUNSHEN XIE 提交于
-
由 gongweibao 提交于
- 26 9月, 2020 3 次提交
-
-
由 Chen Weihang 提交于
-
由 joanna.wozna.intel 提交于
- 25 9月, 2020 22 次提交
-
-
由 LielinJiang 提交于
* refine vision models
-
由 Thunderbrook 提交于
* add xpu in heter mode test=develop * BOOST_CONST_GET; PADDLE_THROW test=develop * code style test=develop * code style test=develop * code style test=develop * refine test=develop * refine test=develop * refine test=develop * refine code test=develop
-
由 Chen Weihang 提交于
* try to fix paddle save load test * open paddle save load * replace dataloader * remove dataloader
-
由 Shang Zhizhou 提交于
* [buf fix]:fix unittest test_activation_op error * split long-time unittests to smaller ones * rename some unittests
-
由 zhang wenhui 提交于
* fix unittest_group_norm_op_v2, test=develop * fix unittest_group_norm_op_v2, test=develop
-
由 Chen Weihang 提交于
-
由 chentianyu03 提交于
* modify ci api white list, add annotation * modify ci api white list, add annotation * move gpu_not_white list to wlist.json * add null line
-
由 tianshuo78520a 提交于
* update gcc8 in python3 ci docker * change cuda 10.2 * update cudnn8 * nvidia error cuda10.2-cudnn8-centos6 images * fix third cache
-
由 mapingshuo 提交于
-
- 24 9月, 2020 7 次提交
-
-
由 Kaipeng Deng 提交于
* make places of DataLoader optional. test=develop
-
由 Huihuang Zheng 提交于
As the title, decrease random failure probability for test_parallel_executor_mnist The old code set larger delta when comparing reduce and all reduce, but didn't set all. I added it. On my linux machine, I run 100 times, no failure occurs. In addition, we only saw this random failure on CI two times since I worked. I thought it was rare and I just increased the delta.