- 30 3月, 2022 1 次提交
-
-
由 huzhiqiang 提交于
-
- 29 3月, 2022 28 次提交
-
-
由 Aurelius84 提交于
* [Eager]Add sort-simple-yaml for automatically sort api|backward.yaml * remove it test=document_fix * refine * add more yaml * remove optional * fix infRT CI
-
由 Chen Weihang 提交于
* unify_kernel_build_target * fix dnn kernel failed * fix dnn kernel loss target * fix xpu compile failed
-
由 Jacek Czaja 提交于
* - update of oneDNN to 2.5 * - changes to UT testing onednn verbose * - Update of oneDNN to 2.5.3 * - update onednn to 2.5.4
-
由 Aurelius84 提交于
-
由 王明冬 提交于
-
由 Roc 提交于
* add random routing op add _random_routing api in utils add random routing ut * # This is a combination of 10 commits. # The first commit's message is: add expert count op add ut for expert_count # This is the 2nd commit message: update UT only for cuda # This is the 3rd commit message: fix for rocm # This is the 4th commit message: update ut # This is the 5th commit message: add moe module # This is the 6th commit message: add expert count op add ut for expert_count # This is the 7th commit message: update UT only for cuda # This is the 8th commit message: update ut # This is the 9th commit message: add moe module # This is the 10th commit message: make expert count private * add assign pos op * fix upper num name * add api _assign pos * add ut for assign pos op * update date * add op about moe gate update utils add limit by capacity op add ut for limit_by_capacity add ut for prune_gate_by_capacity add ut for limit_by_capacity add ut for prune_gate_by_capacity * fix for win * fix bugs in test_limit_by_capacity_op * update ut * update for test (timeout) * fix ut * update * update(fix) ut for win * moe apis in incubate * # This is a combination of 10 commits. # The first commit's message is: add expert count op add ut for expert_count # This is the 2nd commit message: update UT only for cuda # This is the 3rd commit message: fix for rocm # This is the 4th commit message: update ut # This is the 5th commit message: add moe module # This is the 6th commit message: add expert count op add ut for expert_count # This is the 7th commit message: update UT only for cuda # This is the 8th commit message: update ut # This is the 9th commit message: add moe module # This is the 10th commit message: make expert count private * add assign pos op * fix upper num name * add api _assign pos * add ut for assign pos op * update date * fix for win * update for test (timeout) * fix ut * update * fix ut for number count * add apis and utils * add gate apis * add moe and grad clip apis * update moe apis * add ops for moe gate * fix * update for base moe layer api * add random routing op add _random_routing api in utils add random routing ut * fix for dygraph * update with ranodm routing * update * fix ut for limit by capacity * update Co-authored-by: Nhlygit66666 <2570058140@qq.com>
-
由 wangxinxin08 提交于
* add elementwise sub and elementwise div in tensorrt op teller * add unittest of elementwise mul, sub and div
-
由 zhangkaihuo 提交于
-
由 tianshuo78520a 提交于
This reverts commit 9c0eaada.
-
由 tianshuo78520a 提交于
This reverts commit b532315d.
-
由 zhangchunle 提交于
* test labels * test=document_fix
-
由 liutiexing 提交于
-
由 0x45f 提交于
* Fix test_reinforcement_learning.py for eager run_program OP * Add comment
-
由 tianshuo78520a 提交于
This reverts commit e77a947e.
-
由 津 提交于
* [Phi] trans logsumexp op * fix bugs * fix bugs * fix bugs * fix bugs * fix bugs * fix bugs * add sig * fix sig bugs * fix sig bugs * fix xpu bugs * fix review bugs * test=develop
-
由 wuyefeilin 提交于
* mv floordiv to phi * mv elementwise_pow to phi * fix as review
-
由 weishengying 提交于
-
由 shiyutang 提交于
* add_module_in_init_ * Update __init__.py * Update __init__.py
-
由 huangjun12 提交于
-
由 zhangyikun02 提交于
-
由 zn 提交于
-
由 zhangyikun02 提交于
-
由 0x45f 提交于
* Use _C_ops.yolov3_loss in eager mode for test_yolov3.py * fix code for test_yolov3_loss_op * remove useless import * Fix dygraph_mode flag
-
由 From00 提交于
-
由 Chen Weihang 提交于
-
由 Aurelius84 提交于
* [Eager]Switch new Eager mode * switch into eager * fix typo
-
由 Jiabin Yang 提交于
-
由 Zhanlue Yang 提交于
[DoubleGrad PR #2] Adjusted logics of GenerateNodeCreationCodes and GenerateForwardDefinition (#41016) * [Refactor] refactored eager_gen.py PR #2 * [DoubleGrad PR #1] Decoupled code generation logics for Dygraph ForwardFunctions and GradNodes * Fixed minor issue * Adjusted logics of GenerateNodeCreationCodes and GenerateForwardDefinition * Fixed issues * Fixed minor issue
-
- 28 3月, 2022 11 次提交
-
-
由 石晓伟 提交于
-
由 hong 提交于
* move momentum, rmsprop to phi; test=develop * update * update * update * update * udpate; test=develop * fix xpu npu bugs; test=develop * fix npu bug; test=develop * fix windows compile error; test=develop * fix windows compile error; test=develop * polish code; test=develop * fix conflict; test=develop * add meshgrid; * update * polish code * polish code; * fix bug * format; remove useless code * fix npu bug * fix bug
-
由 hong 提交于
* update * add forward case * update * update; test=develop * add some grad kernel; test=develop * move gpu kernel; test=develop * update * update; * update test; * fix selected rows bug; * add mix vector include ; * add mixed vector depen; test=develop * add logit grad signature; * polish code * fix bug; * add namespace for abs * revert code * not move softsign * revmove duplate register; * fix softsign bug * polish code * format * format * fix bug * remove cmake dep * add square sqrt selected rows support * update * remove clip norm * add standalone executor sqrt dep * standalone exec denp sqrt * remove sqrt op in cmkaelist * open some case
-
由 seemingwang 提交于
* fix dependency * reduce warnings
-
由 wangguanqun 提交于
* trainer and worker * delete commonsparsetable from gpups * delete vlog * codestyle * delete communicator from gpups
-
由 Haohongxiang 提交于
* add uts for EagerReducer * add more uts * fix bugs * fix bugs * modify * modify uts * fix bugs * update * update * update * solve conflicts and merge * add some other uts * modify time of uts * update * update * update * remove uts of resnet
-
由 王明冬 提交于
-
由 Jiabin Yang 提交于
-
由 JZ-LIANG 提交于
* add FP16 Pass * Support the auto completion of while_op * acc aligned
-
由 Leo Chen 提交于
-
由 chenjian 提交于
* no * fix bugs * fix doc according to review * fix api doc format * fix api doc according to review * fix bug and add unit test * fix record event bug
-