- 14 9月, 2022 20 次提交
-
-
由 Leo Guo 提交于
Migrate scale and scatter to phi, and modify the code style for roi_align_kernel. test=kunlun (#45938)
-
由 Zhang Jun 提交于
* update * update * update
-
由 Charles-hit 提交于
-
由 Chen Weihang 提交于
* add ci rule for blocking legacy kernel registration, test=document_fix * add a test case, test=document_fix * Revert "add a test case, test=document_fix" This reverts commit 239e08678b791aabb8fae48c4b74f2a8861e222d. * polish desc, test=document_fix
-
由 JZ-LIANG 提交于
* bugfix (#45332) * dist embedding support lookup table v1 * add unitest * customize wait_comm * group gradients * bugfix * update program
-
由 ykkk2333 提交于
-
由 Thunderbrook 提交于
-
由 Leo Chen 提交于
-
由 zhangbo9674 提交于
* support bfloat16 for amp_decorate * add check_finite for bf16 * fix bug * add ut * add ut * refine code
-
由 Xinger 提交于
-
由 Chen Weihang 提交于
* fix arm fp16 compile error * polish macro impl
-
由 xiongkun 提交于
* fix select_input with different shape errors: 1. select_input_with_buildin_type directly return non-undefinedvar branch when meeting undefined var 2. the output shape of select_input is inferred from inputs. * reverse the logic in select_input
-
由 sneaxiy 提交于
* fix distributed_fused_lamb nan * remove CUDA_ASSERT
-
由 Charles-hit 提交于
-
由 wenbin 提交于
-
由 Yiqun Liu 提交于
-
由 Xiaoxu Chen 提交于
* add reduce_mean,reduce_sum primitive ops * add ne_p gt_p primitive operators * add ge_p abs_p primitive oparators
-
由 Chen Weihang 提交于
* normalize yaml op label * revert op_compat yaml change * fix prelu and rnn compat problem * replace api by op
-
由 Chenxiao Niu 提交于
-
由 zhangkaihuo 提交于
-
- 13 9月, 2022 17 次提交
-
-
由 niuliling123 提交于
-
由 Leo Chen 提交于
-
由 Charles-hit 提交于
-
由 Charles-hit 提交于
-
由 Charles-hit 提交于
-
由 Charles-hit 提交于
-
由 pangyoki 提交于
* optimize executor log * delete log in new exe * add log for old executor * use LOG_FIRST_N(INFO, 1)
-
由 Charles-hit 提交于
-
由 ronnywang 提交于
-
由 zyfncg 提交于
* clear extra attr of activation op in opmaker * fix syntax bug * fix mkldnn kernel * fix merge conflict * fix bug
-
由 JingZhuangzhuang 提交于
* add softmax infer kernel
-
由 feifei-111 提交于
* support deco * fix deco ast type * arg_str * 1 * support callable deco * code style * codestyle * test_error * fix decos in another file * recover conflict codes
-
由 ykkk2333 提交于
-
由 ykkk2333 提交于
-
由 ziyoujiyi 提交于
* back fl * delete ssl cert * . * make warning * . * unittest paral degree * solve unittest * heter & multi cloud commm ready * . * . * adapt for nn fl-ps
-
由 Ruibiao Chen 提交于
* Enable standalone executor for single-GPU training * Disable CompiledProgram._graph * Fix CI errors
-
由 zhouweiwei2014 提交于
-
- 10 9月, 2022 1 次提交
-
-
由 qipengh 提交于
-
- 09 9月, 2022 2 次提交
-
-
由 zhaoyingli 提交于
* adapt gradient merge * fix op_role * fix strategy
-
由 zhangbo9674 提交于
* add scope cache & reuse * add gc scope for end of each train step * del scope reuse for jit * refine code * test
-