- 13 6月, 2018 9 次提交
-
-
由 yuyang18 提交于
-
由 yuyang18 提交于
-
由 yuyang18 提交于
-
-
由 Qiao Longfei 提交于
fix build on mac
-
由 tensor-tang 提交于
MKLDNN layout: Support for batch norm operator
-
由 qiaolongfei 提交于
-
由 Wu Yi 提交于
small fix
-
由 qiaolongfei 提交于
-
- 12 6月, 2018 17 次提交
-
-
由 typhoonzero 提交于
-
由 fengjiayi 提交于
Fix a multi-thread bug in readers
-
由 yuyang18 提交于
* less_than * cumsum * multiplex * open_recordio_file
-
由 tensor-tang 提交于
add initial memory flag in MB for infer
-
由 tensor-tang 提交于
-
由 guochaorong 提交于
fix bugs in fluid_benchmark
-
由 Qiao Longfei 提交于
Refine prefetch
-
由 Wu Yi 提交于
* wip * use executor.complete to end trainer * fix build * fix build with distribute off * fix typo * fix cmake typo * fix build
-
由 Tao Luo 提交于
refine docs of elementwise_op etc.
-
由 Wu Yi 提交于
* enable docstring checker and fix comments * update * update by comments * fix build
-
由 Luo Tao 提交于
-
由 qiaolongfei 提交于
-
由 guochaorong 提交于
-
由 qingqing01 提交于
* Add normalization operator. 1. Refine the raw norm_op and let it more general to support to normalize Tensor along any axis. 2. There is a bug in l2_normalize API, which lacks sqrt after `reduce_sum`. 3. Use norm_op to refine the l2_normalize API. 4. Fix bug in test_normalization_wrapper.py.
-
由 Yang Yang(Tony) 提交于
* Create dynamic_graph.md * Update dynamic_graph.md * Update dynamic_graph.md * Update dynamic_graph.md * follow comments, add graph * Update dynamic_graph.md * Update dynamic_graph.md
-
由 Tao Luo 提交于
Build: add make before make install to catch up CMakeLists change.
-
由 Lei Wang 提交于
-
- 11 6月, 2018 14 次提交
-
-
由 whs 提交于
* Add slice op. * Remove using from header file and fix doc. * Fix doc * Small fix.
-
-
由 guochaorong 提交于
-
由 Xin Pan 提交于
Make status update thread-safe
-
由 chengduo 提交于
Enable CPU on Parallel executor
-
由 qiaolongfei 提交于
-
-
由 Luo Tao 提交于
-
由 qiaolongfei 提交于
-
由 gongweibao 提交于
-
由 Xin Pan 提交于
The status is updated in the Process() thread and can be checked in another HandleRequest() thread.
-
由 qiaolongfei 提交于
-
由 Luo Tao 提交于
-
由 qiaolongfei 提交于
-