- 12 10月, 2021 21 次提交
-
-
由 Aganlengzi 提交于
* [NPU] modify for model deepfm * [NPU] unit test delete precision control * [NPU] add more unit test * revert elementwise_mul related modification * [NPU] add more unit tests for concat
-
由 zhouweiwei2014 提交于
-
由 0x45f 提交于
* change time to remove static tempfile * delete remove_static_file() function
-
由 Jacek Czaja 提交于
-
由 Tongxin Bai 提交于
* autograd.functional passed pylint checker. * autograd.functional: fix import errors. * autograd.functional: fixed unit tests. * autograd.functional minor format change
-
由 Qi Li 提交于
* [NPU] fix elementwise_mul to support broadcast, test=develop * remove debug files, test=develop * add axis support, test=develop
-
由 Zeng Jinle 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 HydrogenSulfate 提交于
-
由 Qi Li 提交于
* [NPU] add int64 kernel for scale and slice, test=develop * remove int64 for scale, test=develop
-
由 JingZhuangzhuang 提交于
-
由 Haohongxiang 提交于
* fix calling bug of HybridParallelClipGrad * fix bugs of HybridParallelClipGrad * add unittest of pp with HybridParallelClipGrad * fix bugs in mp_layers.py * update * fix bugs in pp_layers.py * update
-
由 LJQ❤️ 提交于
-
由 Zeng Jinle 提交于
This reverts commit 7e60cc63.
-
由 Aurelius84 提交于
* Fix stop_gradient in RunProgramOp * fix reference
-
- 11 10月, 2021 19 次提交
-
-
由 Leo Chen 提交于
* do not use alignedAllocator when cuda has alignment * update test * fix error during multiple process
-
由 danleifeng 提交于
* heterps:add fuse_allreduce op; test=develop * add program_mode in minimize for pslib mode;test=develop
-
由 jakpiase 提交于
-
由 Zeng Jinle 提交于
* add FLAGS_allreduce_record_one_event * add more comments * fix ut * improve coverage * fix ut, improve coverage
-
由 Liu-xiandong 提交于
Add paddle.nn.functional.sparse_attention API 本个PR主要将sparse_attention功能在python层进行了一层封装,OP的主体代码见:#PR35676 此外,对于封装的python 接口,增加了相应的单测。
-
由 jakpiase 提交于
-
由 Zhang Zheng 提交于
-
由 niuliling123 提交于
* Add functor_primitives.h for kernel primtive api * update * move namespace kps * subFunctor init_data * delete InvalidArgumentError
-
由 Sing_chan 提交于
-
由 Sing_chan 提交于
-
由 Yuang Liu 提交于
-
由 zlsh80826 提交于
Sparse tensor core for convolution requires the input channel dimension is 2:4 structed sparse. So we have to mask the input channel dimension for using sparse tensor core
-
由 caozhou 提交于
* add reshard module * fix conflict * update reshard module * update and add unitest * update reshard module and unitest * add more unitests
-
由 yaoxuefeng 提交于
-
由 tianshuo78520a 提交于
-
由 wangxinxin08 提交于
* enhance yolobox plugin
-
由 Qi Li 提交于
* [NPU] fix matmul_v2 and utils.run_check, test=develop * remove debug files, test=develop * fix install_check, test=develop * fix doc, test=develop * fix review comments, test=develop
-
由 Qi Li 提交于
* [NPU] fix set_value, test=develop * fix typo, test=develop * fix typo, test=develop
-
由 Qi Li 提交于
-