- 16 6月, 2021 20 次提交
-
-
由 Jacek Czaja 提交于
* - Draft of implementation of refactoring - compilation fix * - Fixes after review * - Removed unnecessary comment
-
由 Wilber 提交于
-
由 王明冬 提交于
-
由 lilong12 提交于
-
由 lilong12 提交于
-
由 zhiboniu 提交于
-
由 tianshuo78520a 提交于
-
由 Jiangxinz 提交于
-
由 iducn 提交于
* modify reviewer, test=document_fix * modify reviewer, test=document_fix
-
由 Zhou Wei 提交于
-
由 Thunderbrook 提交于
-
由 zhangbo9674 提交于
* new api trunc, test=develop
-
由 Aurelius84 提交于
-
由 wangguanzhong 提交于
* fix output padding conv * add repr unittest for conv
-
由 Jiangxinz 提交于
-
由 Leo Chen 提交于
* add return when input is tensor * fix typo
-
由 ShenLiang 提交于
* add pplayer * add sharedlayerdesc
-
由 Wangzheee 提交于
-
由 王明冬 提交于
-
由 王明冬 提交于
-
- 15 6月, 2021 11 次提交
-
-
由 Zhou Wei 提交于
-
由 Wilber 提交于
-
由 Leo Chen 提交于
* use SparseSoftmaxCrossEntropyWithLogits * fix * test_slice * revert test_slice * add backprob for npu kernel * fix typo * fix ut * fix ut * refine comments * return softmax
-
由 WeiXin 提交于
* Save all the information of 'ParamBase' in 'Layer'. * edit unittest
-
由 王明冬 提交于
-
由 Jiawei Wang 提交于
* add conv3d prototxt * Update conv3d.pbtxt * Update conv3d.pbtxt
-
由 Shang Zhizhou 提交于
* 1, remove layernorm dynamic fp16; 2, let reshape out in dynamic shape * remove useless code
-
由 wawltor 提交于
add the support for the bool in compare ops
-
由 jiangcheng 提交于
* add reduce_sum_op by add self-kernel * set all ReduceKernel MPType for accuracy * add float16 test script which input is integer number * solve reduce sum float16 check_grad problem * solve conflict and change test script for CI * change kernel register for CI * remove all useless template
-
由 zyfncg 提交于
* Add digamma_op and unittest * add digamma_op api * remove special DigammaCudaKernel and correct some docs * remove unused headers * fix api doc error
-
- 14 6月, 2021 1 次提交
-
-
由 Kaipeng Deng 提交于
* add warning log for DataLoader output format imcompatible upgrade. test=develop
-
- 12 6月, 2021 4 次提交
-
-
由 zhiboniu 提交于
* Eliminate numerical differences of LayerNorm; fix LayerNorm Nan Bug while large data input * fix bug while large shape of data input
-
由 Pei Yang 提交于
* add support for trt dynamic shape flatten op * add version restriction * add ut input dynamic shape
-
由 limingshu 提交于
-
由 joanna.wozna.intel 提交于
* Small changes related to BF16 fusion_gru and fusion_lstm * Correct to pass arg by value * Add conditions to rnn op * Correct the spelling mistake * Improving the test with checking activation * Trigger CI
-
- 11 6月, 2021 4 次提交
-
-
由 ShenLiang 提交于
* fix gather shape bug * fix None * fix topo
-
由 feng_shuai 提交于
* fc_elementwise_layer_fuse_pass * fc_ele_layernorm_pass * fc_elementwise_layernorm_pass * fc_elementwise_layernorm_pass_amend
-
由 ronnywang 提交于
-
由 Jacek Czaja 提交于
* - Second fix - fix * - fix
-