- 11 9月, 2019 1 次提交
-
-
由 Yiqun Liu 提交于
* Refine the codes related to fc op. * Add GPU implementation for fc functor. * Apply fc_fuse_pass in GPU inference. test=develop * Change the cmake for fc op. * Change PADDLE_ENFORCE to PADDLE_ENFORCE_EQ. * Add an attribute to set the activation type in fc_op. * Enhance the unittest of fc_op. test=develop * Remove the declaration of FCOpGrad back to the header file. test=develop * Set default value for newly added arguments in test_fc_op. test=develop
-
- 21 5月, 2019 1 次提交
-
-
由 Yibing Liu 提交于
* Add LAMB optimizer * Expose LAMB Optimizer's APIs test=develop, test=document_preview * Cleanup code & doc test=develop, test=document_preview * Update lamb optimizer's formula test=develop
-
- 11 12月, 2018 1 次提交
-
-
由 Yu Yang 提交于
The macro should be defined by compiler rather than by source. test=develop
-
- 16 11月, 2018 1 次提交
-
-
由 Wu Yi 提交于
* wip simplify operator framework * wip * wip * done test=develop * clean test=develop * fix test=develop * fix deps test=develop * fix cpu build test=develop * fix tensorrt build test=develop * fix tests test=develop * fix test=develop * fix cpu build test=develop
-
- 12 2月, 2018 1 次提交
-
-
由 qingqing01 提交于
-
- 10 2月, 2018 2 次提交
- 26 12月, 2017 1 次提交
-
-
由 Luo Tao 提交于
-
- 12 12月, 2017 1 次提交
-
-
由 QI JUN 提交于
There are mainly following fixes: - take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place` - remove `eigen_device` interface in base class `DeviceContext` - remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext` - remove unused `platform::EigenDeviceConverter` - rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL` - rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`
-
- 21 11月, 2017 1 次提交
-
-
由 Yu Yang 提交于
* Support many data types of several operators * SeqConv only support float/double * Revert adagrad
-
- 13 10月, 2017 1 次提交
-
-
由 Abhinav Arora 提交于
* add adam op moment1_out = beta1 * moment1 + (1 − beta1) * grad moment2_out = beta2 * moment2 + (1 − beta2) * grad * grad moment1_hat = moment1_out / (1 - beta1^t) moment2_hat = moment2_out / (1 - beta2^t) param_out = param - learning_rate * moment1_hat / (sqrt(moment2_hat) + epsilon) * fix moment 2 * Adding the Adam optimization operator * Adding more tests for Adam op
-
- 07 8月, 2017 1 次提交
-
-
由 dongzhihong 提交于
-
- 04 8月, 2017 1 次提交
-
-
由 liaogang 提交于
-
- 31 7月, 2017 1 次提交
-
-
由 qijun 提交于
-
- 25 7月, 2017 1 次提交
-
-
由 Yu Yang 提交于
Make implement an operator less noisy.
-
- 19 7月, 2017 1 次提交
-
-
由 Qiao Longfei 提交于
* a simplest SGD op
-