Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
14,724 Commits 24 Branches 0 Tags 3.9 GiB
Commit Graph

599 Commits

Author SHA1 Message Date
mindspore-ci-bot d6dbfd6593 !9089 [GraphKernel] Add atomic add support 
From: @tronzhang
Reviewed-by: 
Signed-off-by:
 2020-12-03 09:01:46 +08:00
mindspore-ci-bot b15504f6a5 !9327 Cast half to int by rounding down 
From: @jonwe
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
 2020-12-03 05:25:01 +08:00
mindspore-ci-bot 6945c4dd5f !9338 [MS][DynamicShape] Bug in InferImpl for UnsortedSegmentOps (incorrect identification of dynamic mode) 
From: @danishnxt
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
 2020-12-03 04:19:44 +08:00
danishnxt bf5ceaae57 updated to test files + DynamicOpCheckFix 
lintfix

Adding Dynamic_shape_depends
 2020-12-02 13:44:46 -05:00
tom__chen ee8a060f55 Fix unique dynamic shape 2020-12-02 07:49:56 -05:00
tronzhang 2190da9946 support atomic clean and change package for akg. 2020-12-02 16:26:41 +08:00
mindspore-ci-bot eb7cbd49ac !8595 Add Cholesky op at GPU back-end 
From: @peixu_ren
Reviewed-by: @zichun_ye,@wilfchen
Signed-off-by:
 2020-12-02 10:10:24 +08:00
jonwe e2e6bf59b2 Cast half to int by rounding down 2020-12-01 11:48:30 -05:00
mindspore-ci-bot 01df2341ff !9245 Fixes some typos and formatting errors in docstring. 
From: @peilin-wang
Reviewed-by: @robingrosman,@robingrosman,@tom__chen
Signed-off-by: @tom__chen
 2020-12-01 20:08:18 +08:00
mindspore-ci-bot 1ffecf1874 !9180 fix dnnl binary broadcast 
From: @zhao_ting_v
Reviewed-by: 
Signed-off-by:
 2020-12-01 14:52:32 +08:00
peixu_ren af6ded28e3 Add Cholesky op at GPU back-end 2020-11-30 20:27:22 -05:00
zhaoting 102b74682b fix dnnl binary broadcast 2020-12-01 09:20:12 +08:00
Peilin Wang e4444a1c12 initial commit 
fix testcases

fix ci
 2020-11-30 17:52:07 -05:00
mindspore-ci-bot ccf637f64a !9241 SequenceMask gpu kernel 
From: @peilin-wang
Reviewed-by: @tom__chen
Signed-off-by:
 2020-12-01 06:49:47 +08:00
Peilin Wang f7dc2432a0 initial commit 
fix ci

fix ci

remove old sequence mask api

fix ci

fix ci

remove old seuqence mask tests
 2020-11-30 14:44:22 -05:00
mindspore-ci-bot 5835095e07 !9134 Equal op dynamic shape 
From: @jonwe
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @tom__chen
 2020-12-01 00:55:27 +08:00
mindspore-ci-bot dc62360eed !9140 [MS][GPU][CUDA][DynamicShape] - New GPU kernel -> UnsortedSegmentMin + DynamicShape support changes to API + inferImpl func (+SegMax ST correction) 
From: @danishnxt
Reviewed-by: 
Signed-off-by:
 2020-12-01 00:39:08 +08:00
mindspore-ci-bot 2370828043 !8586 fix GatherD and add Randperm for aicpu 
From: @yanzhenxiang2020
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @wuxuejian
 2020-11-28 16:54:13 +08:00
danishnxt 450f1ee886 First Commit - UnsortedSegMin Kernel + DynamicSupport 
lint fix + SegMAX st fix

removing unreq whitspace

lint fix

lint fix2
 2020-11-27 18:29:30 -05:00
mindspore-ci-bot 3e2eb70fc0 !8669 Split op extending PrimitiveWithCheck 
From: @tom__chen
Reviewed-by: 
Signed-off-by:
 2020-11-28 06:50:40 +08:00
mindspore-ci-bot 442217314d !8861 [MS][GPU][DynamicShapeUpdate] Converting UnsortedSegmentMax from normal to dynamic shape op 
From: @danishnxt
Reviewed-by: 
Signed-off-by:
 2020-11-28 06:49:39 +08:00
mindspore-ci-bot 28772d4fb7 !9132 Fix bug for SparseGatherV2 GPU and add dynamic shape testcase 
From: @TFbunny
Reviewed-by: 
Signed-off-by:
 2020-11-28 06:04:20 +08:00
mindspore-ci-bot 9b0ec824c4 !9096 DynamicShapeOp int64 support and bug fix 
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
 2020-11-28 06:01:12 +08:00
mindspore-ci-bot 2e31899101 !9095 [MS][DynamicShape][API] Fixes for UnsortedSegementSum DynamicShape support API + InferImpl func 
From: @danishnxt
Reviewed-by: 
Signed-off-by:
 2020-11-28 05:58:07 +08:00
mindspore-ci-bot b76c7cc7bf !9025 ReLU gpu op supports int32 and int64 
From: @jonwe
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @tom__chen
 2020-11-28 05:54:11 +08:00
tom__chen 4fa6238dde change Split op to extend PrimitiveWithCheck 2020-11-27 15:33:20 -05:00
jonwe 6ae8573803 Equal op dynamic shape 2020-11-27 14:59:53 -05:00
TFbunny 419f8bf72a fix dynamic shape and add testcase for SparseGatherV2 GPU 2020-11-27 12:45:29 -05:00
mindspore-ci-bot f28b73f231 !9040 ExpandDims dynamic shape 
From: @jonwe
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
 2020-11-27 23:41:20 +08:00
mindspore-ci-bot 28e6c7f29e !9034 [MS][DynamicShape] - Converting P.Square to DynamicShape op 
From: @danishnxt
Reviewed-by: @robingrosman
Signed-off-by:
 2020-11-27 23:25:45 +08:00
mindspore-ci-bot b0496aaa10 !8673 Add min/max_shape to ScatterAdd/Update and Transpose and add new dynamic shape testcases 
From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by:
 2020-11-27 22:13:32 +08:00
Jonathan Yan 9f70ebac64 Cast/ReLU dynamic shapelsgpu op supports int32 and int64 2020-11-27 07:03:50 -05:00
jonwe 36ea519009 ExpandDims dynamic shape 2020-11-27 06:24:04 -05:00
mindspore-ci-bot 40222f59a7 !9043 Add support to op L2Loss on gpu 
From: @yuan_shen_zhou
Reviewed-by: @liangchenghui,@liangchenghui
Signed-off-by: @liangchenghui,@liangchenghui
 2020-11-27 17:05:42 +08:00
danishnxt cfc3542261 First Commit SegMax changes + typo in SegSum InferImpl 
UnsortedSegmentMax update + new STs

switch to lambda function

SegmentSum - InferImpl min/max shape pushback fix

lint fix

revert segSum due to sep pr + addressing comments:
 2020-11-26 22:06:52 -05:00
Peilin Wang 1dd302ae93 int64 support and typo fix 
fixed typo

fix pylint
 2020-11-26 20:47:20 -05:00
zhouyuanshen f6e87143c6 add support to op L2Loss on gpu 2020-11-27 09:34:47 +08:00
mindspore-ci-bot 0aa63f21c5 !9039 register FastGelu for activation 
From: @shibeiji
Reviewed-by: 
Signed-off-by:
 2020-11-27 09:10:09 +08:00
danishnxt 6f64fffdb4 Adding API fix for handling dynamic_shape correctly plus STs for SegSum dyn_shape 
Updating InferImpl function

adding check to inferImpl function

lint

lint2
 2020-11-26 18:50:36 -05:00
TFbunny 5e19a642f9 fix and add testcase for dynamic shape scatteradd/update transpose 2020-11-26 17:33:49 -05:00
mindspore-ci-bot e6ebb310eb !8928 [MS][GPU][CUDA] - New GPU kernel -> LinSpace 
From: @danishnxt
Reviewed-by: 
Signed-off-by:
 2020-11-27 05:26:59 +08:00
danishnxt 64a6769d3f Update Square Op to support Dynamic Shape 2020-11-26 12:49:15 -05:00
mindspore-ci-bot c3b057bd69 !8825 [MS][GPU] GatherV2_Dynamic_BugFix 
From: @danishnxt
Reviewed-by: 
Signed-off-by:
 2020-11-27 01:07:32 +08:00
mindspore-ci-bot ebef1df00b !8994 split dropout op and expand dropout 
From: @zengzitao
Reviewed-by: 
Signed-off-by:
 2020-11-26 20:31:51 +08:00
yanzhenxiang2020 082cb4e7fd add Randperm/GatherDGrad and fix GatherD for aicpu 2020-11-26 16:58:20 +08:00
danishnxt a17f76dd1d Initial Commit - GPU LinSpace 
comment fix

docString fix

added asserts in test file atop np checks

lint

lint-2

lint3
 2020-11-26 02:33:55 -05:00
shibeiji cd850bd210 register activation operator fast_gelu 2020-11-26 14:55:39 +08:00
zengzitao 3ef0e9f053 substitute dropout by cudnnuniformreal and dropout 2020-11-26 10:36:02 +08:00
danishnxt 241c8f3d96 GatherUpdate 
test file finishing pending

Update to GatherV2_Bug_Fix

lint fix

lint fix - 2

lint fix

Update to GatherV2 - fixed default inferImpl func + CudeStreamSync

lint fix

SyncDevice

added dynamic shape init_size input

lint
 2020-11-25 17:49:27 -05:00
yanzhenxiang2020 dca109c9a5 fix example of categorical and rnntloss 2020-11-25 20:25:20 +08:00