Commit Graph

5702 Commits

Author SHA1 Message Date
i-robot 311e59c190 !17297 dynamic_shape_pipeline
Merge pull request !17297 from wanyiming/dynamic_shape_pipeline
2021-06-26 02:33:10 +00:00
i-robot 8e043090be !18472 Implement UNet3d on GPU
Merge pull request !18472 from likesen/master
2021-06-25 09:51:46 +00:00
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 b9c178e6b7 !16616 GPU index_add op remove cuda device sync
Merge pull request !16616 from tom_chen/index_add
2021-06-24 14:42:03 +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
i-robot 26e34dec80 !18803 Sync enter bugfix for fail over
Merge pull request !18803 from ZPaC/sync-from-enter
2021-06-24 12:25:52 +00:00
ZPaC 59c23700ab Sync enter bugfix for fail over 2021-06-24 15:34:15 +08:00
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
i-robot 2d8e44f3d0 !18730 add parallel operator for Conv2dTranspose and Conv2DBackPropInput
Merge pull request !18730 from yangzhenzhang/add-parallel-op-for-conv2d-backprop-input
2021-06-24 06:34:30 +00:00
i-robot 6d254ac409 !18679 call nnacl implement for matmul and batchmatmul on ARM
Merge pull request !18679 from zhangbuxue/call_nnacl_implement_for_matmul_and_batchmatmul_on_ARM
2021-06-24 06:14:22 +00:00
i-robot 6bd66b3d70 !18729 Add register limit constraint
Merge pull request !18729 from lishanni/for_regis_limit
2021-06-24 03:48:17 +00:00
yangzhenzhang 69acf757d0 add parallel op for conv2d backprop input 2021-06-24 10:01:49 +08:00
buxue f3dea76b25 call nnacl implement for matmul and batchmatmul on ARM 2021-06-24 09:29:45 +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
Harshvardhan Gupta 6845a16b44 fix coredump with large max_rowsize and add arg validations 2021-06-23 09:31:11 -04:00
lishanni513 8c98146c76 Add register limit constraint 2021-06-23 17:39:22 +08:00
likesen 99a995b432 Implement UNet3d on GPU 2021-06-23 16:22:03 +08:00
i-robot 0c360ea2d6 !16766 Add recompute fuse
Merge pull request !16766 from lingyunli63/recompute_fuse
2021-06-23 07:54:12 +00:00
YangLuo 4e9f6d54f7 update cpp api & doc 2021-06-23 11:14:52 +08:00
tom__chen 799cb79873 remove synchronous error check 2021-06-22 15:10:41 -04:00
i-robot dca5504fd4 !15672 [assistant][RGB2BGR]
Merge pull request !15672 from QingfengLi/rgb2bgr
2021-06-22 12:48:04 +00:00
lingyunli63 a995bea507 recompute_fuse 2021-06-22 20:39:11 +08:00
i-robot 914d77e1c9 !18662 infer optimize to support multithreads and find exit of branches.
Merge pull request !18662 from lanzhineng/infer_optv3
2021-06-22 07:10:01 +00:00
i-robot 9dfc43cf32 !18537 fix type error in profiler
Merge pull request !18537 from yanghaitao/yht_fix_vm_type
2021-06-22 06:40:01 +00:00
wanyiming b37c99b112 dynamic 2021-06-22 13:34:10 +08:00
lanzhineng fc8ec7fc49 infer_optv3:infer optimize to find exit of branches and resolve the
endless issue.
2021-06-22 11:55:56 +08:00
i-robot 972dc627d4 !18660 dataset: open tdt st
Merge pull request !18660 from ms_yan/tdt_st
2021-06-22 03:16:01 +00:00
i-robot edc1c8bf58 !18594 Sync from enterprise
Merge pull request !18594 from ZPaC/sync-from-enter
2021-06-22 02:44:08 +00:00
i-robot 6137a449ed !18462 [MS][GPU] CropAndResize- New GPU Op
Merge pull request !18462 from danishfarid/cropAndResizeGPU
2021-06-21 15:22:05 +00:00
i-robot 64b6f901a6 !18560 ReduceProd gpu support
Merge pull request !18560 from Peilin/reduce-prod-gpu-kernel
2021-06-21 14:42:09 +00:00
ms_yan 750271fa12 open tdt data transfer st 2021-06-21 20:24:30 +08:00
i-robot 4de30f8ecb !18527 [Pynative] Fix bug of nullptr in ms function
Merge pull request !18527 from JoyLvliang/fix_bug_of_nullptr_in_ms_function
2021-06-21 17:25:23 +08:00
yanghaitao1 d2c01a2ff1 fix vm type error 2021-06-21 04:53:49 -04:00
chenx2ovo 4b98099dcb [feat][assistant][I3CEGQ] add op rgb2bgr 2021-06-21 16:16:50 +08:00
i-robot 504ffc6292 !18546 add parallel operator for maxpool and avgpool
Merge pull request !18546 from yangzhenzhang/add-parallel-op-for-maxpool
2021-06-21 14:23:21 +08:00
joylvliang b0bb58a492 fix_bug_of_nullptr_in_ms_function 2021-06-21 11:36:53 +08:00
i-robot 6fb3981170 !18557 Revert the PR of GCN gpu version
Merge pull request !18557 from chengang/revert_gpu_gcn
2021-06-21 10:17:43 +08:00
yangzhenzhang 24370b5613 add parallel op for maxpool 2021-06-21 09:00:56 +08:00
ZPaC 9a8a7c215d Sync from enterprise 2021-06-20 17:18:20 +08:00
i-robot 35b15bda0b !18465 clean code for memory reuse
Merge pull request !18465 from laiyongqiang/mem_clean
2021-06-19 10:12:09 +00:00
i-robot 2a9ab1c211 !18486 upgrade_ascend_0617
Merge pull request !18486 from mindspore_ding/upgrade_ascend_0617
2021-06-19 09:12:18 +00:00
dingpeifei 0af87e888e upgrade_ascend_0617 2021-06-19 16:46:31 +08:00
Peilin Wang ade26aa59f ReduceProd gpu kernel initial commit
update testcase

fix ci

fix ci
2021-06-18 17:01:32 -04:00
danishfarid cf2590c0eb firstCommit gpu CropAndResizeSupport 2021-06-18 11:51:14 -04:00
cristoval b35993e05d Revert "Support GCN on GPU"
This reverts commit 5941d4867c.
2021-06-18 19:36:23 +08:00
i-robot f6cda391a9 !18487 remove the redundant code, add docstring of operator init and add default valuefor args.
Merge pull request !18487 from wangshuide/wsd_master_warn
2021-06-18 09:38:26 +00:00
i-robot ec999d3fa5 !18477 Modify the error message of abstract join
Merge pull request !18477 from huangbingjian/abstract_join
2021-06-18 09:10:29 +00:00
i-robot eb0bad4ad7 !16378 Export bprop func_graph to mindir
Merge pull request !16378 from YuJianfeng/bprop_mindir
2021-06-18 08:50:25 +00:00
huangbingjian d695554ae5 modify the error message of abstract join 2021-06-18 14:12:10 +08:00