- 26 5月, 2020 5 次提交
- 25 5月, 2020 11 次提交
-
-
由 Chen Weihang 提交于
* Revert "Fix test_bert on GPU (#24692)" This reverts commit 62222bf4. * fix random bug in bert unittest, test=develop
-
由 Leo Chen 提交于
-
由 chalsliu 提交于
* Add pr checker for ut test=develop * Change approvers for ut spec checking test=document_fix * Change approvers for ut spec checking test=develop test=document_fix * Add comments test=develop test=document_fix * Refine error message test=develop test=document_fix
-
由 hutuxian 提交于
* Support AucRunner in PaddleBox * update some code style
-
由 Bai Yifan 提交于
-
由 Chen Weihang 提交于
* polish reader error message, test=develop * fix detail error, test=develop * reset activation dcudnn change, test=develop
-
由 wangchaochaohu 提交于
-
由 Liufang Sang 提交于
* remove pow in speed up in dequantize_log test=develop * remove pow in speed up in dequantize_log test=develop * fix unittest test=develop
-
由 Leo Chen 提交于
* support NHWC in Pool2D, test=develop * add unittest, test=develop * fix unittest, test=develop * fix typo, test=develop * follow comments, test=develop * refine comments, test=develop
-
由 tianshuo78520a 提交于
-
由 liym27 提交于
DataLoader makes the data diff even if the data of reader is the same on CUDA place. This PR doesn't use DataLoader to pass the test. we will use DataLoader back after we fix it.
-
- 22 5月, 2020 3 次提交
-
-
由 ShenLiang 提交于
-
由 Leo Chen 提交于
-
由 Jacek Czaja 提交于
-
- 21 5月, 2020 10 次提交
-
-
由 liym27 提交于
[Dy2Stat]Support to transform sequence assignments and multi-target assignments to normal assignments (#24643)
-
由 Adam 提交于
* Update isCached() to be thread freindly test=develop * Add isCached() check inside Softmax handler test=develop * Fix PaddleEnforce() message test=develop
-
由 iducn 提交于
-
由 Guo Sheng 提交于
test=develop
-
由 zhangchunle 提交于
-
由 FDInSky 提交于
* test=develop update interpolate & upsample api
-
由 Huihuang Zheng 提交于
As discussed with QA, we will use p4 machine for unit test and the GPU on those machine may not have enough GPU, which can cause "test_parallel_executor_seresnext_base_gpu" failed. So I decrease the batch size.
-
由 Pei Yang 提交于
-
由 Leo Chen 提交于
-
由 zlsh80826 提交于
-
- 20 5月, 2020 10 次提交
-
-
由 Leo Chen 提交于
* fix bug of varbase.__getitem__, test=develop * fix bug of float and other type, test=develop
-
由 jinyuKING 提交于
API/OP(sequence_first_step, sequence_last_step, sequence_mask, beam_search, beam_search_decode) error message enhancement (#24590)
-
由 iducn 提交于
-
由 Chen Weihang 提交于
-
由 Wilber 提交于
-
由 Jinhua Liang 提交于
-
由 Huihuang Zheng 提交于
The background is that buddy_allocator_test depends on some big GPU allocations and current machine may not have it. It can cause random failure on cloud machine. After discussion with Shuo Tian, we decided to make buddy_allocator_test exclusive.
-
由 Huihuang Zheng 提交于
In the past, the test_cond will fail with 2% probability and easy to re-produce. Now I re-run 300 times and no failure occurs. The probability of still has the failure is (1 - 2%) ^ 300 ~= 0.00004. We can say the random failure disappears. Maybe someone fixed some bugs in PE.
-
由 Chen Weihang 提交于
-
由 Chen Weihang 提交于
* add scale for static runner outputs, test=develop * fix import relation, test=develop * remove len limit, test=develop
-
- 19 5月, 2020 1 次提交
-
-
由 hutuxian 提交于
* Refactor code for dump_field & dump_param: abstracting the common function in base class. * Support dump randomly & random with lineid * Support specify the random interval, which avoids printing too much logs.
-