Commit Graph

12432 Commits

Author SHA1 Message Date
mindspore-ci-bot 81f610c528 !7806 [MS][LITE][CPU]fix bug of int8 3x3 conv bug
Merge pull request !7806 from fuzhiye/tmp
2020-10-27 20:15:25 +08:00
yuchaojie 0d20b99ad1 update transformer's ms version 2020-10-27 20:09:16 +08:00
bingyaweng 11aa3f6f5f modify the method for calculating kl loss 2020-10-27 20:06:39 +08:00
liangchenghui 8093f770bf Add ReverseV2 op for old backend. 2020-10-27 20:00:31 +08:00
mindspore-ci-bot 7c98803ad6 !7646 eval result only return probs
Merge pull request !7646 from wanghua/master
2020-10-27 19:53:51 +08:00
mindspore-ci-bot 78f795971b !7818 add recursion limit for FindParallelCareNode
Merge pull request !7818 from yangzhenzhang/add-recursion-limit
2020-10-27 19:40:15 +08:00
mindspore-ci-bot fbad1e4de8 !7831 [MSLITE][Develop] fix bug of arm cpu int8 op deconv depthwise
Merge pull request !7831 from yangruoqi713/fix_bug
2020-10-27 19:35:44 +08:00
mindspore-ci-bot d4096d733f !7791 gather should be reset to gatherv2 in gpu implementation files
Merge pull request !7791 from zhouyuanshen/master
2020-10-27 19:34:25 +08:00
mindspore-ci-bot 08e6ac0b09 !7805 support split ValueList
Merge pull request !7805 from yihuaijie/master
2020-10-27 19:22:52 +08:00
mindspore-ci-bot 7cbf8346c4 !7837 Modify the Mindspore Version in Wide&Deep and Wide&Deep MultiTable README
Merge pull request !7837 from huangxinjing/doc-version
2020-10-27 19:20:02 +08:00
yao_yf 0cfe72cd22 auto parallel dynamic 2020-10-27 19:17:27 +08:00
yangjie159 f732c7352f sync code to support train 2020-10-27 19:05:35 +08:00
mindspore-ci-bot e7a6ae63bc !7813 [MS][LITE][Develop] add new ops for GPU named hswish
Merge pull request !7813 from pengyongrong/stack
2020-10-27 18:58:54 +08:00
ling d4c73e5683 [MSLITE]int8 split bug 2020-10-27 18:57:58 +08:00
huangxinjing 73b7c04c88 Add version 2020-10-27 18:54:59 +08:00
mindspore-ci-bot b7ad26a87a !7822 optimize opencl depthwise
Merge pull request !7822 from wandongdong/master
2020-10-27 18:50:32 +08:00
Yi Huaijie 3102c4ff8d support split ValueList 2020-10-27 17:58:36 +08:00
zhouyuanshen ff7ecebdcd gather should be reset to gatherv2 in gpu implementation files 2020-10-27 17:53:46 +08:00
linqingke 3465c9c400 new add l2normalize_grad gpu ops. 2020-10-27 17:18:45 +08:00
mindspore-ci-bot 23c5001bde !7823 add node dtype cast ops
Merge pull request !7823 from cjh9368/add_node_dtype_cast
2020-10-27 17:14:16 +08:00
kswang 82667c7a57 pynative wait input tensor 2020-10-27 17:13:05 +08:00
mindspore-ci-bot ce0fd1f297 !7803 remove redundant method root() in tree_adapter.h
Merge pull request !7803 from xiaotianci/fix_redundant_root
2020-10-27 17:08:40 +08:00
mindspore-ci-bot 7852adf12c !7817 modify Retruns
Merge pull request !7817 from lijiaqi/return
2020-10-27 17:00:53 +08:00
mindspore-ci-bot aba7c587be !7814 add_func_graph_cache
Merge pull request !7814 from yao_yf/add_func_graph_cache
2020-10-27 16:54:12 +08:00
mindspore-ci-bot 7f98a95352 !7810 modify ckpt env
Merge pull request !7810 from changzherui/ckpt_env
2020-10-27 16:48:38 +08:00
yangruoqi713 1ec30d4b7e [MSLITE][Develop] fix bug of arm cpu int8 op deconv depthwise 2020-10-27 16:36:34 +08:00
zhaoting f23ffa4ae1 add cpu AssignAdd int32 and int64 2020-10-27 16:28:00 +08:00
mindspore-ci-bot a9eedbc0ea !7820 [MSLITE] Fix the bug of MatVecMul fp32
Merge pull request !7820 from zhanyuan/dev
2020-10-27 16:22:42 +08:00
linqingke a084532a6e modify mass beam_search's floordiv and mod to adapt to 310. 2020-10-27 16:14:40 +08:00
xutianchun 2f5fe427eb add ml_face_mnet smoke test 2020-10-27 16:12:47 +08:00
fuzhiye 0d3c16c141 fix bug 2020-10-27 16:08:57 +08:00
Pengyongrong e8b27b323f anewdd new ops named hswish 2020-10-27 01:07:38 -07:00
lizhenyu 45e504a65b [bugfix]remove useless random seed update 2020-10-27 16:05:09 +08:00
mindspore-ci-bot ac3a82006c !7750 Add a simplification pattern to GraphKernel's ArithSimplify.
Merge pull request !7750 from DeshiChen/1026_simplify_mul
2020-10-27 15:59:13 +08:00
mindspore-ci-bot 4c9f75d13c !7769 fix stream for memreuse
Merge pull request !7769 from gukecai/bug
2020-10-27 15:58:38 +08:00
mindspore-ci-bot e3972c0a7c !7801 fix asynrun segmentfault
Merge pull request !7801 from kisnwang/fix-asynrun-segmentfault
2020-10-27 15:58:14 +08:00
cjh9368 dbd606d17c add dtype_cast op between nodes 2020-10-27 15:57:42 +08:00
mindspore-ci-bot 7e3b0e1027 !7790 add more pynative st test_case
Merge pull request !7790 from lvchangquan/transdata
2020-10-27 15:57:39 +08:00
wandongdong 7bbae002f2 optimize opencl depthwise 2020-10-27 00:52:34 -07:00
mindspore-ci-bot 5b28016b4d !7735 fix gpu multinomial seed issue and GRAPH_MODE
Merge pull request !7735 from baihuawei/fixmultinomial
2020-10-27 15:37:07 +08:00
mindspore-ci-bot 18e3265027 !7717 [MSLITE] Modify usage of benchmark parameter 'inputShapes'.
Merge pull request !7717 from wangshaocong/bugfix_master_v2
2020-10-27 15:33:57 +08:00
mindspore-ci-bot a808ed419d !7802 [MSLITE] sigmoid mul fusion
Merge pull request !7802 from zhengjun10/master
2020-10-27 15:30:18 +08:00
mindspore-ci-bot ef435fbfd1 !7804 [MS][LITE][GPU]optimize conv2d h=w=1
Merge pull request !7804 from chenzupeng/master-lite
2020-10-27 15:29:05 +08:00
mindspore-ci-bot 67102a09b4 !7678 add new available gts model and fix a bug in cast operator
Merge pull request !7678 from XianglongZeng/myms
2020-10-27 15:28:59 +08:00
zhanyuan 371de47944 [MSLITE] Fix the bug of MatVecMul fp32 2020-10-27 15:24:40 +08:00
wanghua f09087a68a update tinybert_model.py 2020-10-27 15:23:02 +08:00
mindspore-ci-bot 52717bba31 !7684 [MSLITE] fp16 run test with settable accuracy
Merge pull request !7684 from ling/bug
2020-10-27 15:22:38 +08:00
mindspore-ci-bot 897ce09346 !7812 GPT scripts bug fix
Merge pull request !7812 from wangjun/master
2020-10-27 15:07:01 +08:00
mindspore-ci-bot 7c06a9643c !7811 remove Assign from parallel black list
Merge pull request !7811 from yihuaijie/dev
2020-10-27 15:01:06 +08:00
yangzhenzhang 92d02b7aff add recursion limit 2020-10-27 14:42:41 +08:00