- 23 3月, 2019 2 次提交
- 22 3月, 2019 5 次提交
-
-
由 phlrain 提交于
-
由 phlrain 提交于
-
由 Xin Pan 提交于
test=develop
-
由 xiaolil1 提交于
* Enable INT8 Concat Kernel to improve the performance of MobileNet-SSD. test=develop * Optimize UT format. test=develop * Fix UT file address issue. test=develop * Refine the license year. test=develop * Optimize code for new API. test=develop * Restructure INT8 Concat kernel. test=develop
-
由 chengduo 提交于
* refine fused_all_reduce_op * add unit test in test_parallel_executor_seresnext test=develop
-
- 21 3月, 2019 3 次提交
- 20 3月, 2019 4 次提交
-
-
由 baojun 提交于
* Add softmax_with_cross_entropy_op test=develop * simplify implementation test=develop
-
由 chengduo 提交于
* fuse all_reduce test=develop * add fuse_parameter_groups_size test=develop * Polish code test=develop * Fix travis-ci test=develop * Add SetGroupAccordingToLayers and SetGroupAccordingToGroupSize test=develop * Add SetGroupAccordingToMemorySize test=develop * fix multi_devices_graph test=develop * reset params_grads test=develop * Polish code test=develop
-
由 Wu Yi 提交于
* wip allreduce in op * wip * wip * wip * wip adding test * wip for conflict with mp mode * fix tests test=develop * fix cpu build test=develop * fix travis clang format test=develop * fix cpu build test=develop * update api.spec test=develop * delete comment test=develop * fix cpplint test=develop * fix test=develop * follow comment test=develop * add file test=develop * fix build test=develop * update test=develop * to be compatible with sync_bn, and fix mp mode in develop test=develop
-
由 lujun 提交于
-
- 19 3月, 2019 3 次提交
-
-
由 Zhen Wang 提交于
-
由 zhhsplendid 提交于
test=develop
-
由 Yibing Liu 提交于
test=develop
-
- 18 3月, 2019 2 次提交
- 15 3月, 2019 5 次提交
-
-
由 Xin Pan 提交于
test=develop
-
由 Xin Pan 提交于
test=develop
-
由 视言 提交于
* Add moving average absmax op in quantilize-aware training.
-
由 qingqing01 提交于
* Support Sync Batch Norm. * Note, do not enable it in one device. Usage: build_strategy = fluid.BuildStrategy() build_strategy.sync_batch_norm = True binary = fluid.compiler.CompiledProgram(tp).with_data_parallel( loss_name=loss_mean.name, build_strategy=build_strategy)
-
由 Yibing Liu 提交于
* Impl fp16 compute kernel for slice_op test=develop * Use data() to replace mutable_data()
-
- 14 3月, 2019 3 次提交
-
-
由 sneaxiy 提交于
test=develop
-
由 sneaxiy 提交于
test=develop
-
由 Zeng Jinle 提交于
test=develop
-
- 12 3月, 2019 8 次提交
-
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
由 dengkaipeng 提交于
-
- 11 3月, 2019 5 次提交
-
-
由 minqiyang 提交于
test=develop
-
由 Qiyang Min 提交于
* Remove Desc in Forward Pass * Refactor VarBase * Add dbg info * Only check type in imperative mode * Polish code and support optimizer test=develop * Fix stop gradient problem in PyLayer test=develop
-
由 minqiyang 提交于
test=develop
-
由 wopeizl 提交于
* fix recordio on win test=develop * test=develop * test=develop * fix code style test=develop * test=develop
-
由 sneaxiy 提交于
test=develop
-