- 10 11月, 2017 2 次提交
-
-
由 tensor-tang 提交于
-
由 Siddharth Goyal 提交于
* Fix attribute naming for momentum_op * Fix minor typo in comment * Fix attribute name * Fix names in test_optimizer * Fix python wrapper
-
- 09 11月, 2017 11 次提交
-
-
由 tensor-tang 提交于
-
由 dangqingqing 提交于
-
由 Luo Tao 提交于
-
由 yangyaming 提交于
-
由 Tao Luo 提交于
-
由 yangyaming 提交于
-
由 Yang Yu 提交于
* increament is default inplace
-
由 fengjiayi 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 skeleton for array_to_lod_tensor and lod_tensor_to_array * Add VarType::LoDTensorArray * Add PyBind of LoDTensorArray * Add InferVarType * Add first unittest * Add ut * Add unittest * Add unittest * Add unittests * update * init * add infershape for lod_tensor_to_array_op * compelete array_to_lod_tensor_op * copy data * clean code * clean code * Fix unittest data * fix bugs * fix compile error * Refine TensorToArrayOp * refactor array_to_lod_tensor * Unittest * fix bugs * Fix unittest * Fix unittest * debug * Debug * Fix unittest * Add grad for ops * Debug * Fix a bug * fix a bug * fix a bug
-
由 Yang Yu 提交于
-
由 Yang Yu 提交于
-
由 Yang Yu 提交于
-
- 08 11月, 2017 22 次提交
-
-
由 tensor-tang 提交于
and abandoned deprecated sum API interface
-
由 wwhu 提交于
-
由 yangyaming 提交于
-
由 yangyaming 提交于
-
由 yangyaming 提交于
-
由 Yang Yang(Tony) 提交于
* add fill_constant_batch_size_like_op to rnn h_boot * first commit * merge develop; fix conflict * update to main_program
-
由 typhoonzero 提交于
-
由 typhoonzero 提交于
-
由 Yang Yu 提交于
CompareOp can run on CPU even other operators are running on GPU, since opeatations like comparing control flags should be performed only on CPU
-
由 Yang Yu 提交于
Follow comments
-
由 Yang Yu 提交于
-
由 chengduoZH 提交于
-
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 skeleton for array_to_lod_tensor and lod_tensor_to_array * Add VarType::LoDTensorArray * Add PyBind of LoDTensorArray * Add InferVarType * Add first unittest * Add ut * Add unittest * Add unittest * Add unittests * update * init * add infershape for lod_tensor_to_array_op * compelete array_to_lod_tensor_op * copy data * clean code * clean code * Fix unittest data * fix bugs * fix compile error * Refine TensorToArrayOp * refactor array_to_lod_tensor * Unittest * fix bugs * Fix unittest * Fix unittest * debug * Debug * Fix unittest * clean code * refactor * use ostream * update test * fix gpu build error * make gpu test pass
-
由 Yang Yu 提交于
Used for shrink memories state in DyRNN. The height of state could be shrinked after running a step block.
-
由 Yu Yang 提交于
-
由 Yu Yang 提交于
-
由 Yu Yang 提交于
-
由 Yu Yang 提交于
-
由 Yu Yang 提交于
-
由 Yu Yang 提交于
* Chage `IndicateDataType` to `GetKernelType`. Make it easier to understand. * Change `OpKernelKey` to `OpKernelType` * Make operator developers can customize which kernel the operator will use in runtime.
-
由 Yu Yang 提交于
* Compare Operator * Follow comments
-
由 qingqing01 提交于
-
- 07 11月, 2017 5 次提交
-
-
由 tensor-tang 提交于
-
由 typhoonzero 提交于
-
由 typhoonzero 提交于
-
由 typhoonzero 提交于
-
由 dangqingqing 提交于
1. user can disable peephole connections. 2. not calculate some gradients.
-