Col_lin
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
28,065 Commits 18 Branches 22 Tags 3.6 GiB
Commit Graph

2545 Commits

Author SHA1 Message Date
lichenever a7f8024c29 add_replace_graph_for_conv2d 2021-07-28 11:02:21 +08:00
i-robot b8901526b9 !20882 Fix GradOperation recursive call issues. 
Merge pull request !20882 from 张清华/opt
 2021-07-27 01:00:39 +00:00
Zhang Qinghua 8be269c1cc Fix GradOperation issues: 
- Allow the J CNode used multiple as non-call input.
- Check recursive J FG including FV J CNode.
- Add grad functional api.
 2021-07-26 20:47:49 +08:00
yangzhenzhang 80e5cc0e52 add parallel op for gatherd 2021-07-26 11:14:25 +08:00
i-robot c80098b261 !20712 [Auto parallel] Enable All2All when inferring redistribution ops 
Merge pull request !20712 from Xiaoda/76-change-alltoall-redistribution
 2021-07-24 07:07:16 +00:00
i-robot 26bfbf9124 !18282 [assistant][ops]New operator implementation, include HSigmoid and HSigmoidGrad 
Merge pull request !18282 from 张璇/hsigmoid_grad
 2021-07-24 06:22:46 +00:00
Xiaoda Zhang bb5d4212f7 enable All2All in infering redistribution ops 2021-07-24 09:09:01 +08:00
i-robot 1e1d7d5739 !20443 Add numpy-native ut/st 
Merge pull request !20443 from 杨林枫/add_numpy_ut
 2021-07-23 11:24:39 +00:00
wanglixin 33e93e296c [fix][assistant][I3PYD0] fix bug in the Ascend operator HSigmoid and HSigmoidGrad 2021-07-23 16:33:16 +08:00
wild-fox b6c575689d [feat][assistant][I3PYD0] add new Ascend operator HSigmoid 2021-07-23 16:20:51 +08:00
wanglixin a377ade89f [feat][assistant][I3PYD3] add new Ascend operator HSigmoidGrad 2021-07-23 16:20:51 +08:00
i-robot 0f8a21693b !18266 [assistant][ops]New operator implementation, include SoftShrink and SoftShrinkGrad 
Merge pull request !18266 from 张璇/softshrink
 2021-07-23 06:40:47 +00:00
i-robot 69a7ef8632 !20707 [AutoParallel]AutoParallel support control flow 
Merge pull request !20707 from lichen/auto_parallel_support_control_flow
 2021-07-23 06:22:12 +00:00
i-robot bcb49f10b4 !20702 Fix NumpySlicesDataset not accept NumPy as data 
Merge pull request !20702 from xiaotianci/fix_numpy_slices
 2021-07-23 03:31:29 +00:00
i-robot 7666b6e388 !20708 fix exception type to TypeError 
Merge pull request !20708 from xychow/fix-exception-type
 2021-07-23 02:48:49 +00:00
i-robot e8448f972b !20725 Support the tensor of string dtype. 
Merge pull request !20725 from 张清华/opt
 2021-07-23 02:12:48 +00:00
Zhang Qinghua f91c697efc Support the tensor of string dtype. 2021-07-22 19:05:37 +08:00
lichenever 3c7cfb7c08 auto_parallel_support_control_flow 2021-07-22 17:31:31 +08:00
Xiao Tianci 53c01c437d fix NumpySlicesDataset not accept numpy 2021-07-22 15:46:37 +08:00
wangrao124 463ca2434c add numpy-native ut/st test files 2021-07-22 14:47:00 +08:00
zhousiyi 9319eff958 change the error type from RuntimeError to TypeError 2021-07-22 06:38:12 +00:00
Lizxxy 5d00d482e4 [feat][assistant][I3PYDC] add new Ascend operator SoftShrinkGrad 2021-07-20 17:55:02 +08:00
zx be26e62080 [feat][assistant][I3PYDB] add new Ascend operator SoftShrink 2021-07-20 17:52:46 +08:00
i-robot a7d40fc220 !20520 [AutoParallel]Add op AllToAllv 
Merge pull request !20520 from lichen/add_op_AllToAllv
 2021-07-20 07:57:03 +00:00
i-robot ed4c9682b5 !20505 limit the scope of lift free variable 
Merge pull request !20505 from xychow/limit-lift-scope
 2021-07-20 06:10:17 +00:00
lichenever 8c1998fd6b add_op_AllToAllv 2021-07-20 10:49:28 +08:00
i-robot c9d3c1d346 !20411 enable optimizer parallel for inference 
Merge pull request !20411 from gziyan/enable_opt_shard_predict
 2021-07-20 01:37:06 +00:00
i-robot 6061194083 !20490 update check strategy for conv2d 
Merge pull request !20490 from yangzhenzhang/update-check-strategy-for-conv2d
 2021-07-20 01:23:08 +00:00
zhousiyi 597f29ea7d limit the lambda lift scope to the func_graph passed 2021-07-20 00:37:21 +00:00
zhousiyi b590f6d929 don't replace u with fprop_u in primal_graph and bprop_fg 2021-07-19 08:14:40 +00:00
yangzhenzhang b31cd27a08 update check strategy for conv2d 2021-07-19 15:03:26 +08:00
Ziyan 1c9166e0a6 remove restriction for opt shard in inference 2021-07-16 17:16:12 +08:00
Xiaoda Zhang 04381273b3 Add the sharding propagation function: 
1) users configure sharding strategies for operators;
2) framework will propagate the strategies from configured-ops to
non-configured ops using BFS;
3) the propagation goal is to minimize redistribution communication
cost;
 2021-07-14 10:54:18 +08:00
i-robot ea440db85e !18488 Remove Parameter Check in PyNative mode 
Merge pull request !18488 from caifubi/master-pynative-no-parameter
 2021-07-05 12:05:46 +00:00
i-robot 8508c9eb45 !18179 [MD] add slice patches op in c++ and python 
Merge pull request !18179 from liyong126/slice_patches_op
 2021-07-05 07:04:45 +00:00
i-robot 4a8225eaf4 !19160 Fix parameter check of Rotate and GaussianBlur API 
Merge pull request !19160 from xiaotianci/fix_rotate_and_gaussian_blur
 2021-07-05 06:30:46 +00:00
i-robot b93907fdca !19246 Change Loss to LossBase 
Merge pull request !19246 from chenhaozhe/change-loss-base
 2021-07-05 01:24:23 +00:00
liyong 9a1de0af65 add slice patches in python and c++ 2021-07-05 09:16:47 +08:00
zhousiyi a5b4e7bbf8 lift fv before grad except weight, then convert 
switch(cond, partial(g1, xs), partial(g2, ys))(Zs)
 to
 switch(cond, g1, g2)(Xs, Ys, Zs)

 switch_layer(index, make_tuple(partial(g1, xs), partial(g2, ys)))(Zs)
 to
 switch_layer(index, make_tuple(g1, g2))(Xs, Ys, Zs)

put Zs at last when unifyparameter as it may have u-monad or io-monad

use joined args other than broadened one as some extra parameter which is not a parameter of while_header can be add to while_body

inline fprop_switch forcely

reorder the parameter if one of the parameter is Monad when incorporate call

incorporate switch tuple_getitem if item 0 of tuple is EnvInstance or
item 1 of tuple is bprop function

addn with shape() and shape(1)

remove context_ from FuncGraphEvaluator to make it re-entry able to resolve evaluator stuck issue because of re-entry of the same FuncGraphEvaluator
 2021-07-03 00:03:37 +08:00
chenhaozhe 086a871975 Change Loss to LossBase 2021-07-02 16:41:02 +08:00
Xiao Tianci 8cdd0d8d90 fix param check for Rotate and GaussianBlur 2021-07-01 20:31:33 +08:00
i-robot 8102d68b8b !18420 Give concreted error when mappable dataset lack __len__ method 
Merge pull request !18420 from luoyang/API-check-pybind-name
 2021-06-29 07:48:03 +00:00
YangLuo 17c121b087 Give concreted error when mappable dataset lack __len__ method 2021-06-29 11:13:12 +08:00
lichenever db8850a4a3 pipeline_support_predict_master 2021-06-26 09:25:59 +08:00
i-robot 905cba5ae0 !18788 Fix cache_admin stderr message and TFRecord get_dataset_size() when cached 
Merge pull request !18788 from lixiachen/fix_stderr_msg
 2021-06-24 17:52:10 +00:00
i-robot 708e56f659 !17899 Fix compile cache bug for resent50 
Merge pull request !17899 from LiangZhibo/cache
 2021-06-24 13:29:58 +00:00
Ziyan be1f5a43d7 opt shard fit micro batch 2021-06-24 14:52:19 +08:00
l00591931 8ae5d7cc84 Fix compile cache for resnet50 2021-06-24 14:36:24 +08:00
yangzhenzhang 69acf757d0 add parallel op for conv2d backprop input 2021-06-24 10:01:49 +08:00
Lixia Chen 1f03b16e0e Fix cache_admin server start stderr output 
Fix TFRecord get_dataset_size() when cached
 2021-06-23 18:32:54 -04:00