Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
0
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
42,838 Commits 24 Branches 0 Tags 3.9 GiB
Commit Graph

558 Commits

Author SHA1 Message Date
mindspore-ci-bot 3d06cbf987 !4801 Must set or change parallel mode before any Initializer created 
Merge pull request !4801 from yihuaijie/dev
 2020-08-21 09:47:32 +08:00
Yi Huaijie 89a4ebf8a1 parallel mode must be set before create an initializer 2020-08-20 19:20:46 +08:00
mindspore-ci-bot 9ee144ea40 !4744 [AutoParallel]Support bert 
Merge pull request !4744 from lichen/support_bert
 2020-08-20 14:13:19 +08:00
lichenever 221a801395 auto parallel support bert 2020-08-19 17:47:59 +08:00
yangzhenzhang cda08f6a52 concat 3 tensors in auto parallel mode 2020-08-19 11:27:35 +08:00
mindspore-ci-bot 2ae6365d77 !4650 EmbeddingLookup support auto parallel 
Merge pull request !4650 from yangzhenzhang/embedding-lookup-auto-parallel
 2020-08-18 15:51:04 +08:00
yangzhenzhang 6f6a8ae9f0 embedding lookup auto parallel 2020-08-18 11:40:09 +08:00
Yi Huaijie 0f7ead5f14 parameter slice init test all initializers 2020-08-18 10:34:15 +08:00
yao_yf cbb4363fa7 remove to_full_tensor and load_inputs in exexute stage 2020-08-14 11:40:27 +08:00
yangzhenzhang 14c77c9f03 update field split 2020-08-14 09:00:56 +08:00
mindspore-ci-bot 2db0290c49 !4356 Add validation for field split 
Merge pull request !4356 from yangzhenzhang/update-field-split
 2020-08-13 16:43:44 +08:00
yangzhenzhang 4a0e6ff7fc update field split 2020-08-13 10:35:04 +08:00
yao_yf e4de26d5bc embeddinglookup wrap 2020-08-12 14:24:13 +08:00
yangzhenzhang f4bb43bbaf add concat op 2020-08-07 15:47:17 +08:00
lichenever bfc96de1b9 add dropout distributed op 2020-08-05 14:07:24 +08:00
simson 3617121ccf revert modification of opt 2020-08-04 10:09:21 +08:00
Xiaoda Zhang d24a902afe add a new graph operation in autoparallel 2020-07-31 10:57:19 +08:00
mindspore-ci-bot ab4c43007f !3657 Add parallel operator for StridedSlice 
Merge pull request !3657 from yangzhenzhang/add-stridedslice-op
 2020-07-30 15:02:35 +08:00
Ziyan 98e2ee90de fix optimizer parallel problems 2020-07-29 20:55:31 +08:00
yangzhenzhang 9aa84b3d14 add strided slice op 2020-07-29 17:18:50 +08:00
mindspore-ci-bot 16079e6356 !3472 [Auto parallel] Cost model for GPU 
Merge pull request !3472 from Xiaoda/13-add-gpu-costmodel
 2020-07-28 09:27:26 +08:00
mindspore-ci-bot d4165671d9 !3435 Add parallel operator for Tile 
Merge pull request !3435 from yangzhenzhang/add-tile-op
 2020-07-28 09:16:30 +08:00
Xiaoda Zhang 9097b36950 add resnet50 testcases for gpu 2020-07-27 15:22:30 +08:00
lirongzhen1 51796aa624 fix sparse feature bug for auto parallel 2020-07-27 10:44:49 +08:00
yangzhenzhang 6a6e2bd271 add tile op 2020-07-25 11:42:06 +08:00
panyifeng 963bd67a60 add sparse api docs 2020-07-24 15:40:46 +08:00
panyifeng 8a89f003eb fix sparse related issues 2020-07-23 19:05:53 +08:00
mindspore-ci-bot 684ff4f46b !3160 Rewrite tensor's __bool__ for pynative mode 
Merge pull request !3160 from Simson/push-to-opensource
 2020-07-23 16:55:49 +08:00
simson 5f77fbdd75 Rewrite tensor's __bool__ for pynative mode 2020-07-23 16:51:46 +08:00
mindspore-ci-bot 7f1ccc5f3b !3311 add sparse feature test cases for auto parallel 
Merge pull request !3311 from lirongzhen1/master
 2020-07-23 11:35:49 +08:00
mindspore-ci-bot bc20de741a !3315 restore reshape ut 
Merge pull request !3315 from yao_yf/restore_auto_parallel_reshape_ut
 2020-07-22 16:23:29 +08:00
yao_yf 1d3a06a3b0 recover reshape ut 2020-07-22 14:30:12 +08:00
lirongzhen1 5d63c60135 add sparse feature test cases for auto parallel 2020-07-22 14:27:12 +08:00
wangnan39@huawei.com 082433183d uniform learning_rate behavior of optimizers 2020-07-20 21:09:20 +08:00
anzhengqi 008b91b2a1 inject epoch ctrl op in the execution tree and send eos at the end of epoch 2020-07-20 13:02:47 +08:00
liuxiao93 75881e5f2f check input of BatchNorm is 4D. 2020-07-18 14:53:11 +08:00
wangnan39@huawei.com 86889c59cb optimizer adapt IndexedSlices 2020-07-17 15:57:50 +08:00
yao_yf abebb2004b remove 4 reshape ut 2020-07-17 10:56:24 +08:00
mindspore-ci-bot edec821c50 !2876 set reshape operator no redistribution for auto parallel 
Merge pull request !2876 from lirongzhen1/master
 2020-07-17 10:01:06 +08:00
mindspore-ci-bot 74bbfa3cf6 !3095 modify the limit of loss scale 
Merge pull request !3095 from Simson/push-to-opensource
 2020-07-16 21:30:54 +08:00
lirongzhen1 c1eba79b83 set reshape redistribution strategy attribute to no redistribution 2020-07-16 21:20:46 +08:00
simson 177e18f3f4 modify the limit of loss scale 2020-07-16 16:44:02 +08:00
Ziyan 39f08eb7dd enable optimizer parallel 2020-07-16 11:47:41 +08:00
lichenever cde5cc2bd2 add_embedding_look_up 2020-07-13 22:00:55 +08:00
Xiaoda Zhang fc906f7f58 move embeddinglookup to external 2020-07-08 16:27:40 +08:00
Yi Huaijie cae254f4df asymmetric row split support for GatherV2 2020-07-07 14:36:07 +08:00
jinyaohui dd5fba1db9 add notice 2020-06-29 17:39:25 +08:00
Ziyan 0925e35252 enable optimizer parallel with broadcast 2020-06-27 14:36:17 +08:00
Ziyan 41ddc153a6 modify lars interface 2020-06-23 11:23:28 +08:00
Xiaoda Zhang 3ff6e336c6 check cast from optimizer in auto-parallel 2020-06-22 09:10:22 +08:00
mindspore-ci-bot c0fe8c0322 !2273 [AutoParallel]update EmbeddingLookUp op 
Merge pull request !2273 from lichen/fix_embeding_look_up
 2020-06-18 21:07:52 +08:00
lichenever 563622874a update 2020-06-18 19:45:40 +08:00
Xiaoda Zhang 69574f3823 fix the bprob error of embeddinglookup 2020-06-18 16:06:56 +08:00
Xiaoda Zhang 55e7d9d2b8 move embeddinglookup to the internal 2020-06-17 14:11:28 +08:00
Xiaoda Zhang 20d2012a0e implementing the backward of embeddinglookup 2020-06-13 16:57:56 +08:00
lichenever e0e055a0b8 add sparse gatherv2 2020-06-12 11:09:28 +08:00
mindspore-ci-bot 5b0472683c !1737 sparse feature backpropagation 
Merge pull request !1737 from lirongzhen1/sparse
 2020-06-12 09:51:41 +08:00
Yi Huaijie e5c351690b support load full dataset on each device 2020-06-11 19:48:06 +08:00
lirongzhen1 516b56cb64 sparse feature bp 2020-06-11 16:32:24 +08:00
Xiaoda Zhang 1cfb52bc0e add the reshape part of the embeddinglookup backward operator 2020-06-03 10:39:45 +08:00
lichenever 1437966c98 gatherv2_support_host_and_device 2020-06-01 21:06:35 +08:00
Xiaoda Zhang 55ef468a7f fix the embeddinglookup bug 2020-06-01 15:52:22 +08:00
Yi Huaijie d7e1ab1445 clean pylint warnings 2020-05-30 11:35:00 +08:00
jinyaohui 86d197dfeb clean pylint 2020-05-29 23:00:00 +08:00
mindspore-ci-bot 8316f736ea !1650 Support forward reduce scatter for Matmul 
Merge pull request !1650 from yangzhenzhang/forward-reduce-scatter
 2020-05-29 14:45:32 +08:00
mindspore-ci-bot 57874cd61f !1372 [Auto parallel] Add a new primitive EmbeddingLookup 
Merge pull request !1372 from Xiaoda/add-embedinglookup-primitive
 2020-05-29 11:06:25 +08:00
yangzhenzhang 19bd830539 support forward reduce scatter for matmul 2020-05-29 10:48:32 +08:00
mindspore-ci-bot f523a0f83c !1600 [AutoParallel]Fix GatherV2 bug 
Merge pull request !1600 from lichen/fix_auto_parallel_gatherv2_bug
 2020-05-29 10:27:14 +08:00
Xiaoda Zhang 4154adf196 add embedinglookup primitive 2020-05-29 10:15:09 +08:00
lichenever c223fde566 fix auto parallel gatherv2 bug 2020-05-28 20:34:35 +08:00
yangzhenzhang 1413f520d7 support reshape optimized 2020-05-28 16:05:36 +08:00
mindspore-ci-bot 0fe3ff2761 !1506 Setting auto parallel flag for dataset wrap cell 
Merge pull request !1506 from yangzhenzhang/set-auto-parallel-for-dataset-wrap
 2020-05-27 21:50:43 +08:00
yangzhenzhang 2f8516e5d7 set auto parallel for dataset warp cell 2020-05-27 15:27:22 +08:00
yao_yf 96c9569dca fix reshape reshape case 2020-05-27 14:25:40 +08:00
yangzhenzhang 7c237620ba add sigmoid op 2020-05-26 21:33:32 +08:00
Yi Huaijie ac484ebbc0 remove unused variable 2020-05-25 19:59:28 +08:00
Yi Huaijie 0b137e5312 delete unused arguments in test cases 2020-05-25 15:46:05 +08:00
Yi Huaijie 8cfc05e4cf clean pylint warnings of parallel test cases 2020-05-25 11:08:29 +08:00
mindspore-ci-bot e87e6b38b0 !1355 [AutoParallel]Fix GatherV2 distributed op 
Merge pull request !1355 from lichen/fix_gatherv2
 2020-05-23 17:04:41 +08:00
lichenever 390a86effb fix gatherv2 2020-05-23 15:38:50 +08:00
Yi Huaijie 1e6ee83874 delete parallel end-to-end test cases 2020-05-22 17:35:18 +08:00
Yi Huaijie 14fe72f383 fix pylint warnings 2020-05-22 14:54:36 +08:00
jinyaohui fbdba6e4da clean pylint 2020-05-22 13:23:13 +08:00
Yi Huaijie 7a5004cc49 init the slices of a Initialzer on different devices 2020-05-21 20:43:15 +08:00
mindspore-ci-bot 1b7cdc4c61 !1218 clean pylint warning in test about import 
Merge pull request !1218 from jinyaohui/pylint1
 2020-05-18 19:54:04 +08:00
jinyaohui 5a914994ba clean pylint 2020-05-18 16:42:35 +08:00
yangzhenzhang 6b54a6417d ckpt and restore parameter shape 2020-05-18 15:16:29 +08:00
jinyaohui bcfaff97f9 clean pylint 2020-05-18 10:31:46 +08:00
jinyaohui 2907cf4488 remove some context param 2020-05-16 15:07:13 +08:00
candanzg 2cc85bdc93 Support weight compile according to shape 
Signed-off-by: candanzg <zhangshucheng@huawei.com>
 2020-05-13 17:49:07 +08:00
mindspore-ci-bot 0345995000 !1110 [AutoParallel]fix gatherv2 and dataset bug 
Merge pull request !1110 from lichen/fix_gatherv2_and_dataset_bug
 2020-05-13 11:42:09 +08:00
lichenever debfd38b75 fix gatherv2 and dataset bug 2020-05-13 09:44:37 +08:00
jinyaohui 391a060f21 remove two context param 2020-05-12 22:11:50 +08:00
mindspore-ci-bot 48e54dcfda !1056 [Auto parallel] Memory calculation in the inference phase 
Merge pull request !1056 from Xiaoda/memory-estimation-in-inference-phase
 2020-05-12 17:21:05 +08:00
mindspore-ci-bot fdad91355f !1093 reset parallel context before running each parallel test case 
Merge pull request !1093 from yihuaijie/master
 2020-05-12 15:57:55 +08:00
Xiaoda Zhang a05aa21cc2 calculating PEAK memory cost in the inference phase 2020-05-12 15:30:59 +08:00
mindspore-ci-bot 06a9eeb3bf !1050 [bug][auto_mixed_precision]fix sens shape error of `TrainOneStepWithLossScaleCell` 
Merge pull request !1050 from vlne-v1/I1GEPN-check-bprop-error-loss-scale-wrap
 2020-05-12 15:23:46 +08:00
Yi Huaijie 462f30e077 reset auto parallel context before each test 2020-05-12 14:59:45 +08:00
Wei Luning 3e89f7baaa fix sens shape error in loss_scale one_step_wrap 2020-05-12 14:09:51 +08:00
mindspore-ci-bot 2af6ee2482 !1054 Add slice shape for param info 
Merge pull request !1054 from yangzhenzhang/add-slice-shape-for-param-info
 2020-05-11 21:29:40 +08:00
mindspore-ci-bot 57085bb18d !996 [Auto parallel] Supporting VirtualDataset having three inputs 
Merge pull request !996 from Xiaoda/supporting-virtualdataset-has-three-inputs
 2020-05-11 18:59:09 +08:00
yangzhenzhang 05fde3d23d add slice shape for param info 2020-05-11 16:01:55 +08:00
yao_yf 716def7c0a move InferStraByTensorInfo to tensor_info.h 2020-05-11 14:44:06 +08:00
mindspore-ci-bot dd2062bf8d !1023 add_gatherv2_distributed_op 
Merge pull request !1023 from lichen/add_gatherv2_distributed_op
 2020-05-11 09:18:10 +08:00
lichenever 19a24b86ac add gatherv2 distributed op 2020-05-09 17:20:50 +08:00
yao_yf f0bf438a55 reshape strategy search 2020-05-09 17:00:37 +08:00
Xiaoda Zhang 5e041966f1 add a new vritualdataset cell for three inputs 2020-05-09 11:10:33 +08:00
yangzhenzhang 8c9730b3c5 add parallel mode for cell 2020-05-07 20:11:20 +08:00
Xiaoda Zhang def8573275 implementing-searching-strategy-for-inference 2020-05-07 14:53:50 +08:00
yao_yf 5a6540450e use param name as the key of strategy checkpoint 2020-04-30 11:01:32 +08:00
mindspore-ci-bot 21d936e656 !728 auto parallel strategy checkpoint full 
Merge pull request !728 from yao_yf/strategy_checkpoint_extend
 2020-04-28 19:51:15 +08:00
yao_yf 6cde5f6d91 auto parallel strategy checkpoint 2020-04-28 15:48:41 +08:00
zjun c538b83712 remove enbale hccl 2020-04-28 10:37:08 +08:00
mindspore-ci-bot afbd24cb78 !718 Fix dtype judge sentence in infer_dtype function of hcom operations 
Merge pull request !718 from zhouyuanshen/master
 2020-04-27 21:23:25 +08:00
zhouyuanshen c046874b03 fix bug in infer_dtype function of hcom operations 2020-04-27 18:49:21 +08:00
Xiaoda Zhang e227415673 support-the-multiple-subgraphs-in-the-ANF 2020-04-27 09:34:41 +08:00
yangzhenzhang 4750861054 fix layernorm bug 2020-04-26 16:39:32 +08:00
yangzhenzhang 36a62576e8 support forward graph 2020-04-26 14:44:23 +08:00
Xiaoda Zhang 3ff9e54734 add the resnet50 32k-8p testcase 2020-04-26 09:07:37 +08:00
lirongzhen1 56f785f7e6 add context configration 2020-04-24 19:35:14 +08:00
mindspore-ci-bot ce71c17933 !645 auto parallel prelu operator support broadcast 
Merge pull request !645 from yao_yf/auto_parallel_prelu_support_broadcast
 2020-04-24 17:32:37 +08:00
mindspore-ci-bot 84d5e4f923 !643 [AutoParallel]Support reshape parameter 
Merge pull request !643 from lichen/support_reshape_parameter
 2020-04-24 16:43:29 +08:00
mindspore-ci-bot 00859ae119 !586 enable/disable allreduce_fusion 
Merge pull request !586 from lirongzhen1/allreduce_fusion
 2020-04-24 16:02:53 +08:00
lichenever 2ab211ae04 support reshape parameter 2020-04-24 15:59:09 +08:00
yao_yf 425276d43d auto parallel prelu support prelu 2020-04-24 14:10:57 +08:00
Xiaoda Zhang dfde76af88 delete the 'simplify_cal' attribute in 'set_algo_parameters' and 'get_algo_parameters' interface 2020-04-24 10:46:38 +08:00
lirongzhen 4ff418084c enable/disable allreduce_fusion 2020-04-24 10:34:59 +08:00
lichenever c78630d737 support multiple subgraphs 2020-04-22 17:40:56 +08:00
Ziyan 0d208e00bd Model ALLTOALL as a single operator in cost model; scale the ALLTOALL, 
ALLGATHER, and REDUCESCATTER with different factors; change the BETA and
GAMMA value in cost model.
 2020-04-20 09:11:03 +08:00
yangzhenzhang 36ffb66782 add parallel op for square 2020-04-17 10:42:52 +08:00
yangzhenzhang 57cd9f8188 add parallel op for sigmoidloss 2020-04-16 14:33:24 +08:00
yangzhenzhang 6d522f0a4f add parallel op for layernorm 2020-04-15 19:30:07 +08:00
Xiaoda Zhang ffb2cb03a4 Change 'NOT_FULLY_USE_DEVICES' to 'FULLY_USE_DEVICES' and make ALL-1 user-specified-strategy valid in auto-parallel 2020-04-15 10:14:46 +08:00
lichenever b81cc6ea4f add minimum distributed op 2020-04-13 17:26:31 +08:00
mindspore-ci-bot 7bc2cee318 !167 add_squeeze_distributed_op 
Merge pull request !167 from lichen/add_squeeze_distributed_op
 2020-04-13 10:03:25 +08:00
c00425699 c8cdb6b331 support distributed GatherV2 operator 2020-04-12 16:00:47 +08:00
buxue 5841fe010e Support pow's second input could be tensor and fix bug in bprop of pow 2020-04-11 16:16:58 +08:00
lichenever 32cd280c1a add squeeze distributed op 2020-04-11 16:03:08 +08:00
yangzhenzhang b34c0e7a17 add parallel op for dropoutdomask 2020-04-11 12:11:22 +08:00
yao_yf b5e3fa9593 fix auto parallel prelu 2020-04-08 20:45:08 +08:00
yangzhenzhang dd0d4e6b84 add parallel ops for expand dims 2020-04-03 15:17:07 +08:00
mindspore-ci-bot a5a904fbdf !91 fix bug for allreduce fusion and add resnet unit test 
Merge pull request !91 from chentingting/allreduce_fusion_resnet
 2020-04-02 21:35:35 +08:00
mindspore-ci-bot 55916351ee !52 remove ge depend 
Merge pull request !52 from flywind/remove-ge-depend
 2020-04-02 20:59:06 +08:00
Wei Luning 73ba399364 remove ge depend in cpu 2020-04-02 19:38:42 +08:00
c00425699 ab917a734d fix bug for allreduce fusion and add resnet unit test 2020-04-02 18:55:04 +08:00
lichenever 5240b1f603 fix refkey bug for auto parallel 2020-04-02 16:14:29 +08:00
mindspore-ci-bot a47046652a !76 [Auto parallel] Refining the strategy_checking for resnset50 
Merge pull request !76 from Xiaoda/modify-resnet50-strategy-checking
 2020-04-01 16:36:13 +08:00
mindspore-ci-bot 22a9c00bcd !57 Add parallel operators for Neg and BatchMatMul 
Merge pull request !57 from yangzhenzhang/master
 2020-04-01 15:23:46 +08:00
Xiaoda Zhang fb6eed23ae refining strategy-checking for resnet50 2020-04-01 14:58:58 +08:00
mindspore-ci-bot 87040483ee !58 fix two cast bug in auto parallel 
Merge pull request !58 from lichen/fix_two_cast_bug_in_auto_parallel
 2020-04-01 09:54:10 +08:00
yangzhenzhang 110640e2ad add parallel ops for neg and batchmatmul 2020-04-01 09:36:38 +08:00
mindspore-ci-bot e2df848597 !55 modify long time python ut 
Merge pull request !55 from changzherui/master
 2020-03-31 21:01:40 +08:00
lichenever 2da38ad401 fix two cast bug in auto parallel 2020-03-31 18:54:21 +08:00
chang zherui eaf7146d46 modify longtime python ut 2020-03-31 16:18:04 +08:00
lichenever f946aea10d fix grpah mode loop sink bug in auto parallel 2020-03-31 16:16:53 +08:00
leonwanghui 976af212e9 回退 'Pull Request !17 : [AutoParallel]Fix bug in the case of two cast' 2020-03-31 10:27:40 +08:00
lichenever b4d34973bc fix_cast_bug 2020-03-30 21:13:45 +08:00
zhunaipan 930a1fb0a8 initial version 
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
 2020-03-27 22:54:54 +08:00