- 17 4月, 2019 3 次提交
-
-
由 Hongyu Liu 提交于
Fix bpr loss
-
由 Hongyu Liu 提交于
separate runtime infershape
-
由 Hongyu Liu 提交于
fix shape check many by hongyu
-
- 16 4月, 2019 37 次提交
-
-
由 SunGaofeng 提交于
modify infer shape pad_op, pad_constant_like_op, no need in psroi_pool_op, crop_op
-
由 tangwei12 提交于
* fix infershape in runtime * fix infershape in runtime test=develop * fix infershape in runtime
-
由 Hongyu Liu 提交于
softmax cross entropy support high rank
-
由 XiaoguangHu 提交于
polish yolov3_loss annotation
-
由 xiaoting 提交于
-
由 guru4elephant 提交于
move split filelist from trainer.py to fleet & fix error
-
由 Kaipeng Deng 提交于
Fix yolo param
-
-
由 phlrain 提交于
-
由 Qiyang Min 提交于
Imperative tracer does not hold op any more
-
由 jerrywgz 提交于
refine infer shape in roi align
-
由 xiaoting 提交于
-
由 tensor-tang 提交于
test=develop
-
由 Hongyu Liu 提交于
Rsqrt
-
由 Hongyu Liu 提交于
Zeros like
-
由 tensor-tang 提交于
test=develop
-
由 tensor-tang 提交于
test=develop
-
由 Tao Luo 提交于
Reduce trt ci times.
-
https://github.com/PaddlePaddle/Paddle由 phlrain 提交于
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into softmax_cross_support_high_rank
-
由 phlrain 提交于
-
由 liuwei1031 提交于
* scatter_op bug fix, test=develop * scatter_op bug fix, test=develop * scatter_op bug fix, test=develop
-
由 jerrywgz 提交于
fix batch_norm and cos_sim shape check
-
由 Hongyu Liu 提交于
modified infer shape for op
-
由 tianshuo78520a 提交于
* test=develop * test=develop * test=develop
-
由 zhoukunsheng 提交于
* test=develop update API.spec with hash, resolve conflict * remove unused parameter
-
由 wopeizl 提交于
-
https://github.com/PaddlePaddle/Paddle由 phlrain 提交于
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into softmax_cross_support_high_rank
-
由 Tao Luo 提交于
Fix merge_lod_tensor_op infer shape
-
由 phlrain 提交于
-
由 Tao Luo 提交于
Reduce the layer number of transfromer model
-
由 Tao Luo 提交于
Fix bug: long vgg-utest testing time
-
由 jerrywgz 提交于
-
由 Hongyu Liu 提交于
Fix concat shape
-
由 zhaoyuchen2018 提交于
Fix sum infershape issue
-
由 whs 提交于
test=develop
-
由 乔龙飞 Qiao Longfei 提交于
fix split_byref_op infer shape
-
由 Jacek Czaja 提交于
* - Reuse of conv PD - conv transpose pd reused - Added PD reusing of softmax and Batch Norm - Refactoring and removal of not needed routines of mkl-dnn ops test=develop - Fix to reusing conv test=develop - Lint fixes test=develop - Further lint fixes test=develop - Lint fixes test=develop - lint fixes test=develop - Lint workaround test=develop * - Fix after review on including boost as third party header test=develop * - Fix after review. Name change to something more descriptive test=develop
-