- 03 4月, 2019 2 次提交
-
-
由 Qiyang Min 提交于
* Fix bug of gradient interface * shrink transformer * Right transformer * Change from width-first backward to deep-first backward process test=develop * Reverse iterator op's input test=develop * Polish code * Change the iteration direction in ingrads' map slots test=develop * Polish code test=develop * Add GPU place in static mode of ptb rnn test=develop * Polish code test=develop
-
由 Zeng Jinle 提交于
* fix some grad op desc maker test=develop * fix grad op desc makers test=develop
-
- 02 4月, 2019 7 次提交
-
-
由 Xin Pan 提交于
test=develop
-
由 chengduo 提交于
* Fix the bug of the fast threaded executor. I
-
由 Zeng Jinle 提交于
test=develop
-
由 gongweibao 提交于
-
由 Qiyang Min 提交于
* Fix bug of gradient interface * shrink transformer * Right transformer * Change from width-first backward to deep-first backward process test=develop * Reverse iterator op's input test=develop * Polish code * Change the iteration direction in ingrads' map slots test=develop * Polish code test=develop
-
由 Jiabin Yang 提交于
* test=develop, add FC and test * test=develop, refine code
-
由 AIFollowers 提交于
-
- 01 4月, 2019 5 次提交
-
-
由 Jiabin Yang 提交于
-
由 Genieliu 提交于
-
由 dongdaxiang 提交于
test=develop
-
由 Yan Xu 提交于
* add seresnet unittest test=develop * add dropout layer test=develop * fix ci test=develop * fix comment test=develop * fix comment test=develop * fix ci test=develop * fix ci test=develop * fix ci * fix module name test=develop * run imperative serenext unit test serially test=develop
-
由 chengduo 提交于
test=develop
-
- 31 3月, 2019 3 次提交
-
-
由 lujun 提交于
-
由 qingqing01 提交于
* Add linear learning warmup method This warmup lr can be combinated with other learning rate strategies. For example: decayed_lr = fluid.layers.linear_lr_warmup( fluid.layers.piecewise_decay(boundaries, lr_steps), warmup_steps, start_lr, end_lr) -
由 Wu Yi 提交于
* fix cudnn batch norm accuracy test=develop * fix cudnn batch norm accuracy test=develop * disable failed test for later fix test=develop
-
- 30 3月, 2019 4 次提交
-
-
由 Qiao Longfei 提交于
-
由 dongdaxiang 提交于
test=develop
-
由 xjqbest 提交于
test=develop
-
由 chengduo 提交于
* Add DecoupledWeightDecay
-
- 29 3月, 2019 19 次提交
-
-
由 lujun 提交于
-
由 dongdaxiang 提交于
test=develop
-
由 lujun 提交于
-
由 whs 提交于
test=develop
-
由 Qiao Longfei 提交于
-
由 minqiyang 提交于
test=develop
-
由 peizhilin 提交于
-
由 wopeizl 提交于
* add slice support for dim < 0 test=develop
-
由 dongdaxiang 提交于
test=develop
-
由 minqiyang 提交于
test=develop
-
由 lujun 提交于
-
由 minqiyang 提交于
test=develop
-
由 dongdaxiang 提交于
test=develop
-
由 Genieliu 提交于
1 metric update should be in the for loop. 2 sample_num is the shape's first element 3 preds cast should use round, astype("int32") method will cast all the elements(which is less than 1) to 0. -
由 lujun 提交于
-
由 AIFollowers 提交于
* Add a tool to summary model's PARAMS, FLOPs in paddle/fluid/contrib.
-
由 dongdaxiang 提交于
test=develop
-
由 dongdaxiang 提交于
test=develop
-
由 dongdaxiang 提交于
test=develop
-