- 04 11月, 2017 4 次提交
-
-
由 Qiao Longfei 提交于
* add acc layer * memory log level change from 3 to 10 * use gaussian random to init conv parameters * use initializer * fix import * batch_norm use helper to create persistable var * refine code * train only 2 batches for test * use g_program and g_init_program * use XavierInitializer to init fc parameter
-
由 Yu Yang 提交于
* Add LoDRankTable LoD Rank Table stores the `level` of `lod` which is ordered by sequence length in descending order. It is useful when implement dynamic RNN and is shared by dynamic RNN memory, dynamic RNN slice input and dynamic RNN slice output operators. * Add InferVarType
-
由 Abhinav Arora 提交于
-
由 Kexin Zhao 提交于
-
- 03 11月, 2017 3 次提交
-
-
由 Yi Wang 提交于
-
由 Kexin Zhao 提交于
-
由 Yu Yang 提交于
* Fix bug in lookup_table_op & layers * Missing Act in layers * Should += in CPU * Remove check in python * Fix bug in sequence_conv_pool() * Fix a bug in test_recommender_system.py * Just skip test_evaluator
-
- 02 11月, 2017 10 次提交
-
-
由 dangqingqing 提交于
-
由 Yang Yang(Tony) 提交于
-
由 Yu Yang 提交于
* Init commit * Make executor use ProgramDescBind * Change Attribute from BlockDesc to BlockDescBind * Since we will get the program desc in RNN, just BlockDesc is not enough. * Add DeviceContext to Executor API * Rewrite RNN * Pass Python * AddBiasOp does not care num_flatten_dims * Stash * Fix MacOS Compile * Pass RNN forward * add python test * refactor test * Make compile pass * add gradopmaker * First draft done * Polish code * add grad op maker and grad infershape * Polish code * Fix backward.cc bug * Fix infershape * Rename function * add backward test * simplify recurrent test * Update * Pass unittest * Add comments & refine test * Add comments * refactor test * Complete Unittest * fix StepScopes enforce * Remove unused unittest * no type error * Update * Make RNN Pass unittest
-
由 Yang Yang(Tony) 提交于
-
由 zchen0211 提交于
-
由 zchen0211 提交于
-
由 Yang Yang(Tony) 提交于
-
由 zchen0211 提交于
-
由 dangqingqing 提交于
-
由 dangqingqing 提交于
-
- 01 11月, 2017 10 次提交
-
-
由 yangyaming 提交于
-
由 zchen0211 提交于
-
由 dzhwinter 提交于
* "replace enum with string" * "fix layers"
-
由 Yu Yang 提交于
* Init commit * Make executor use ProgramDescBind * Change Attribute from BlockDesc to BlockDescBind * Since we will get the program desc in RNN, just BlockDesc is not enough.
-
由 Qiao Longfei 提交于
* add shareLod * fix sequence_conv grad infershape
-
由 QI JUN 提交于
-
由 fengjiayi 提交于
1. Change some `auto` to `auto*` 2. Change `Tensor` to `LoDTensor`
-
由 Abhinav Arora 提交于
-
由 武毅 提交于
* refine evaluator op types * update * follow comments * update * fix v2 mnist case * fix v2 mnist case * update * update
-
由 zchen0211 提交于
-
- 31 10月, 2017 13 次提交
-
-
由 caoying03 提交于
-
由 caoying03 提交于
-
由 dangqingqing 提交于
-
由 dangqingqing 提交于
-
由 dangqingqing 提交于
-
由 dangqingqing 提交于
Use posix_memalign to allocate aligned memory, since some SIMD instructions require the alignment of memory accesses.
-
由 武毅 提交于
* refine evaluator op types * update * follow comments * update * fix v2 mnist case * fix v2 mnist case * update * update
-
由 caoying03 提交于
-
由 zchen0211 提交于
-
由 zchen0211 提交于
-
由 fengjiayi 提交于
* Complete accuray layer * Fix error * Fix error * Add 'accuracy' to __all__ * update * Fix Type error * Fix error * Refine unit tests * Fix an unit test error
-
由 QI JUN 提交于
* reimplement pow operator * add pow_grad operator * fix code style * fix build error * fix op_test bug * revert pow operator * add FIXME comment
-
由 fengjiayi 提交于
* Fix Type error * Fix error * Fix top_k_op GPU code data type
-