Commit Graph

19139 Commits

Author SHA1 Message Date
buxue 610300335f support index by negative number 2021-02-26 12:49:25 +08:00
mindspore-ci-bot ec439fdb97 !12503 refactor the c++ code of gpu profiler
From: @gzhcv
Reviewed-by: 
Signed-off-by:
2021-02-26 11:45:24 +08:00
jiangzhenguang 8a7b85ec24 amend deeplabv3 export.py 2021-02-26 11:26:04 +08:00
zuochuanyong 73cdfa6086 modify graph info to fix bug of sync stream error 2021-02-25 22:00:53 -05:00
mindspore-ci-bot 2191fa3a2e !12626 Fix some error format of api comments.
From: @zhang_yi2020
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
2021-02-26 10:51:00 +08:00
mindspore-ci-bot 8cc861beca !12642 Clear memory in singlel op graph after pre build
From: @HulkTang
Reviewed-by: @chujinjin,@zhoufeng54
Signed-off-by: @chujinjin
2021-02-26 10:40:01 +08:00
mindspore-ci-bot dc538edfe9 !12635 numpy-native fix docstring for trunc
From: @jachua
Reviewed-by: @liangchenghui,@guoqi1024
Signed-off-by: @liangchenghui
2021-02-26 10:11:20 +08:00
mindspore-ci-bot 03272a3999 !12625 modify CheckpointStrategy to adapt load operator
From: @huangbingjian
Reviewed-by: @hwhewei,@zh_qh
Signed-off-by: @zh_qh
2021-02-26 10:11:09 +08:00
mindspore-ci-bot de1367d82c !12333 Add grad impl for op MatrixDiag/MatrixDiagPart/MatrixSetDiag
From: @yuan_shen_zhou
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
2021-02-26 09:31:35 +08:00
mindspore-ci-bot 0cc784b49f !12439 GPU fused_batchnorm_ex op dynamic shape support
From: @tom__chen
Reviewed-by: 
Signed-off-by:
2021-02-26 09:03:51 +08:00
mohammad f73def5933 make C++ API classes for samplers 2021-02-25 17:29:43 -05:00
Eric d328d6eb54 Adding affine
Need to move Affine to image_utilC

Refactored affine

Added affine API

Added API class impl

Removed default from vision ir

Compiling

removed extra file

Spell check
2021-02-25 16:55:02 -05:00
mindspore-ci-bot a99fcf890a !12437 [Ascend][GPU] Trigger final execution graph dump on dump enabled
From: @islam_amin
Reviewed-by: @john_tzanakakis
Signed-off-by:
2021-02-26 04:42:16 +08:00
TFBunny ec644a5291 add fix to float64 support in reduce ops 2021-02-25 15:30:58 -05:00
mindspore-ci-bot c8bca27340 !12604 add shape check for sequence_mask
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
2021-02-26 02:02:11 +08:00
Islam Amin 722eb2ec5a ascend graph dump trigger at data dump 2021-02-25 12:04:59 -05:00
mindspore-ci-bot c8defc2a8a !12558 Fix sampler error messages
From: @hfarahat
Reviewed-by: 
Signed-off-by:
2021-02-26 00:10:17 +08:00
mindspore-ci-bot 7dc13edc2e !12554 [MD] Support C++ API class for 3 data transforms ops and some code cleanup
From: @tina_mengting_zhang
Reviewed-by: 
Signed-off-by:
2021-02-26 00:09:19 +08:00
mindspore-ci-bot 7e78eb4550 !12643 Revert 'Pull Request !12569 : Add circle check in ub fusion'
From: @xsmq
Reviewed-by: @lilongfei15,@zhunaipan
Signed-off-by:
2021-02-25 23:22:47 +08:00
gongxiaoqing 4d6ff51ca1 回退 'Pull Request !12569 : Add circle check in ub fusion' 2021-02-25 22:28:37 +08:00
tom__chen 03b1aeecdb FusedBatchNormEx dynamic shape support 2021-02-25 09:03:40 -05:00
tanghuikang 4faadcfec6 Clear memory in singlel op graph after pre build 2021-02-25 21:51:20 +08:00
dayschan 7beca18f3c Refactor GraphKernelExpander (3rd submission)
Rewrited the OpInfer as a class with functions "infer_shape","infer_type"
and "infer_format". Used op name to find the subclass that have implemented
these functions. Two common class "_Elemwise" and "_Reduce" was provided.

Op BiasAddGrad supported "FractalNZ" format.
2021-02-25 19:14:46 +08:00
huangmengxi b29937c453 fix trunc docstring 2021-02-25 18:57:40 +08:00
mindspore-ci-bot 9ab234369b !12616 [MS][LITE][BUG] fix static check bug
From: @cjh9368
Reviewed-by: @zhanghaibo5,@hangangqiang
Signed-off-by: @hangangqiang
2021-02-25 18:55:01 +08:00
gzhcv c3a9fe6d06 refactor the c++ code of gpu profiler
Some function from ccsrc/profiler/device/cpu/cpu_data_saver* and ccsrc/profiler/device/gpu/data_saver* could be reused. Some function from ccsrc/profiler/device/cpu/cpu_profiling* and ccsrc/profiler/device/gpu/gpu_profiling* could be reused. This pr is to simplify these code.
2021-02-25 17:47:07 +08:00
zhangxinfeng3 1bd849462c Add bond, angle and dihedral modules of Sponge 2021-02-25 17:30:39 +08:00
Jiaqi 112b8488c5 add rmse loss 2021-02-25 17:03:07 +08:00
zhangyi e4000470cf fix error format of docstrings for some api comments. 2021-02-25 15:47:30 +08:00
mindspore-ci-bot 29b68a82a4 !12602 update print tensor_type to align to run package
From: @yepei6
Reviewed-by: @kisnwang,@kingxian
Signed-off-by: @kingxian
2021-02-25 14:38:18 +08:00
chenhaozhe 4a1fc91cca update interface in ascend310_infer of SSD 2021-02-25 14:28:12 +08:00
mindspore-ci-bot 34daed0fbe !12608 change dimension of input for FusedBatchNormEx from 2D to 4D in test_two_matmul_batchnorm_ex.
From: @wangshuide2020
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
2021-02-25 14:06:43 +08:00
huangbingjian 0bbd95d7a0 modify CheckpointStrategy to adapt load operator 2021-02-25 11:57:49 +08:00
cjh9368 638d330fc7 [MS][LITE] solve static check bug 2021-02-25 11:22:13 +08:00
mindspore-ci-bot 52fac12367 !12592 fix checkpoint cache and prosess exception
From: @fangzehua
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54
2021-02-25 11:01:47 +08:00
mindspore-ci-bot 50d45e746c !12595 kernel build server process support to print log
From: @zhoufeng54
Reviewed-by: @kisnwang,@xu-yfei
Signed-off-by: @xu-yfei
2021-02-25 10:59:50 +08:00
mindspore-ci-bot 5498f9ef8a !12582 910 parallel inference
From: @zhoufeng54
Reviewed-by: @kisnwang,@xu-yfei
Signed-off-by: @xu-yfei
2021-02-25 10:58:57 +08:00
mindspore-ci-bot a5523ffbb4 !12494 [MS][LITE]support lite mindrt
From: @zhang_xue_tong
Reviewed-by: 
Signed-off-by:
2021-02-25 10:20:28 +08:00
l00591931 65cbb7b08f Change io_format in adapter 2021-02-25 10:14:24 +08:00
wangrao ab17c49eba add sinh, cosh, asinh, acosh, atanh, atan2, asinhgrad, acoshgrad for cpu 2021-02-25 10:07:21 +08:00
liuyu 994261e409 add master new cops 2021-02-25 10:06:43 +08:00
yujianfeng 41189781f3 support scalar input for cell 2021-02-25 09:56:15 +08:00
wangshuide2020 72e938eb06 change dimension of input for FusedBatchNormEx from 2D to 4D in test_two_matmul_batchnorm_ex. 2021-02-25 09:37:59 +08:00
mindspore-ci-bot 5b206557d2 !12586 support invert bool tensor
From: @zhangbuxue
Reviewed-by: @zh_qh,@ginfung,@zh_qh
Signed-off-by: @zh_qh,@zh_qh
2021-02-25 09:37:43 +08:00
mindspore-ci-bot cf8279267c !12468 [MSLITE][Develop] fix bug of npu kernel fusion
From: @yangruoqi713
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
2021-02-25 09:35:48 +08:00
mindspore-ci-bot b5ad38fab8 !12600 modify the processing method of isolated nodes
From: @huangbingjian
Reviewed-by: @zh_qh,@hwhewei
Signed-off-by: @zh_qh
2021-02-25 09:30:10 +08:00
mindspore-ci-bot 9147f57e40 !11317 modify dropout unify mindir pass
From: @yuchaojie
Reviewed-by: @jjfeing,@kisnwang
Signed-off-by: @kisnwang
2021-02-25 09:29:03 +08:00
mindspore-ci-bot eeb7291d51 !12543 add StridedSliceGrad op
From: @lyvette
Reviewed-by: 
Signed-off-by:
2021-02-25 09:18:53 +08:00
mindspore-ci-bot 750fa6417c !12598 fix bug of conv2d cpp infer
From: @lianliguang
Reviewed-by: @ginfung,@chujinjin
Signed-off-by: @chujinjin
2021-02-25 09:16:35 +08:00
mindspore-ci-bot 08f8570338 !12569 Add circle check in ub fusion
From: @irmo
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
2021-02-25 09:12:23 +08:00