Commit Graph

45876 Commits

Author SHA1 Message Date
i-robot 2e43226628
!34077 [MS][LITE][parallel] lite support share packed weight
Merge pull request !34077 from yefeng/306_fix_pack_weight
2022-06-06 06:32:05 +00:00
buxue d8d0b1b55b fix coredump for mindir fuzz test 2022-06-06 14:24:11 +08:00
Xiaoda Zhang 3d074d7281 add renorm cpu implementation; vmap for renorm 2022-06-06 14:23:43 +08:00
i-robot 3362365868
!35407 bugfix for embedding cache
Merge pull request !35407 from zyli2020/master
2022-06-06 06:23:23 +00:00
r1chardf1d0 21113f1788 batch_to_space_nd 2022-06-06 14:21:57 +08:00
i-robot 93d22454c1
!35192 Refactor MKL base kernel && adapt new int64_t API
Merge pull request !35192 from zhujingxuan/logsoftmax
2022-06-06 06:08:41 +00:00
i-robot f5814afc26
!35122 support is_dynamic_impl
Merge pull request !35122 from liubuyu/dynamic
2022-06-06 06:02:26 +00:00
i-robot d0272fe908
!35453 takedown testBGCF_amazon_beauty to ensure gate stability
Merge pull request !35453 from yanghaoran/takedown_testcases
2022-06-06 05:46:18 +00:00
liyuanshuo123@yeah.net fd0a70042b [feat][assistant][I48O8I] add new ascend operator ListDiff 2022-06-06 13:39:21 +08:00
wilfChen b519d4ac20 priority replay buffer primitive c 2022-06-06 12:15:47 +08:00
Erpim 9cddd36024 fix vmap multi-nested tuple input bug 2022-06-06 11:54:41 +08:00
chenmai1102 16e74974f3 support S string input in GE mode 2022-06-06 11:48:12 +08:00
zhengzuohe 5b2701bbfe udpate docs for functional api of ScatterUpdate 2022-06-06 11:44:35 +08:00
dayschan 90d026dfbb Op ScatterNd, rewrite infer function with c++, supports vmap, supports dynamic shape on Gpu and Ascend. 2022-06-06 11:44:33 +08:00
yanghaoran bd303abb5b takedown testBGCF_amazon_beauty to ensure gate stability 2022-06-06 11:24:34 +08:00
zhangfanghe 5e64f86c3f optimizes the documentation of ApplyMomentum. 2022-06-06 11:16:14 +08:00
i-robot 27cb08f4f3
!34976 AdamApplyOne support dynamic shape
Merge pull request !34976 from huoxinyou/0525adam
2022-06-06 02:49:21 +00:00
i-robot f57a593863
!35221 [MS][LITE][parallel predict] add model for parallel predict: GPU
Merge pull request !35221 from yefeng/330-add_model_for_parallel_predict_GPU
2022-06-06 02:34:23 +00:00
i-robot 76799d0f63
!34630 align to 512 btyes
Merge pull request !34630 from zhupuxu/fix_bug_for_clean_size
2022-06-06 02:31:52 +00:00
yefeng 8540f98405 lite support packed weight share 2022-06-06 10:22:40 +08:00
ttudu 6ad90c4e1e print_reduce pass gpu in dynamic_shape 2022-06-06 10:21:57 +08:00
i-robot 3743ca1504
!35265 [MS][LITE][parallel predict] parallel predict UT
Merge pull request !35265 from yefeng/332-add_UT_for_parallel_predict_api
2022-06-06 02:19:30 +00:00
i-robot 03d8dd39b5
!35352 Modify the infererence function and functional interface of ScatterUpdate
Merge pull request !35352 from zhengzuohe/scatterupdate
2022-06-06 01:55:25 +00:00
i-robot 99942cb39e
!35344 Add col2im and csr_sparse_matrix_to_sparse_tensor operator primitive interface.
Merge pull request !35344 from liangchenghui/add_csr_primitive
2022-06-06 01:48:03 +00:00
i-robot f4f153d3c3
!35427 fixed the controlflow bug in dynamic feed mode
Merge pull request !35427 from huoxinyou/control_flow_bug
2022-06-06 01:44:54 +00:00
i-robot 5f0048104b
!35384 custom aicpu use host mem
Merge pull request !35384 from zhoufeng/custom-aicpu-use-host-mem
2022-06-06 01:32:59 +00:00
i-robot a933e0e9fd
!35364 [MS]{Lite][Task] move less_kernel_mock to kernel dir
Merge pull request !35364 from 刘力力/feature_cloud_infer_runtime_develop3
2022-06-06 01:30:47 +00:00
hezhenhao1 92239f01ad Opt GPU implementation of NonZero operator. 2022-06-06 09:18:14 +08:00
xuanyue c2e58df4a6 mul-reduce fusion and matmul-opt for specific scene 2022-06-06 09:14:58 +08:00
i-robot 5f08b78e2c
!35024 Dynamic Shape Support for SparseApplyRMSProp Op
Merge pull request !35024 from alashkari/sparse_apply_rms_prop_dynamic_shape
2022-06-05 17:54:51 +00:00
danishfarid 722e88995a UpsampleTrilinear3D added
changed to new base class for GPU

working changes + checks

final updates + lint fixes

code check fixes

lint fix 2

lint fix 3

lint fix 4

lint fix final

fix

minor fix

minor fix 2

removed merge marker

address comments

commit fix

comments addressed
2022-06-05 16:52:36 +00:00
i-robot 9e06255cef
!31784 [assistant][ops] Add new Outer
Merge pull request !31784 from wxy220/outer
2022-06-05 15:27:19 +00:00
gyuer 0ae2540d9a [feat][assistant][I4XJG4] add new functional Ldexp 2022-06-05 21:12:31 +08:00
huoxinyou f1a14f89fa setitem & getitem support dynamic shape 2022-06-05 19:47:48 +08:00
liangchenghui 7b0ea2e0e5 Add col2im and csr_sparse_matrix_to_sparse_tensor operator primitive interface. 2022-06-05 13:58:39 +08:00
fanjibin 938a17d314 add intopk vmap, functional/tensor api 2022-06-05 13:18:51 +08:00
huoxinyou 5c16f8a24d fixed the controlflow bug in dynamic feed mode 2022-06-05 10:23:42 +08:00
limingqi107 5916b3c90b runtime performance optimize-add the pass of MultiActorFusion 2022-06-05 10:14:50 +08:00
i-robot e29229320e
!35284 [MD] Adding docstrings for minddata UT Python Part 4
Merge pull request !35284 from davidanugraha/add_dataset_test_comment_part4
2022-06-04 17:46:20 +00:00
i-robot f058997021
!35071 Add feature ReflectionPad1d & ReflecitonPad2d
Merge pull request !35071 from shaojunsong/feature/reflectionPad
2022-06-04 13:00:45 +00:00
shaojunsong c0b9d74987 added reflectionpad1d and 2d 2022-06-04 17:43:49 +08:00
lizhenyu 0a441c0198 bugfix for embedding cache 2022-06-04 16:25:41 +08:00
ZPaC 302d1dbbb5 bypass rank id and size 2022-06-04 12:02:53 +08:00
polyhedral 92b1459d57 flatten support functional and tensor interface, dynamic shape and vmap 2022-06-04 11:49:55 +08:00
minara 2446919c63 Trunc op commit 2022-06-03 08:45:54 -04:00
zy efdb47b0a1 [feat] [assistant] [I40GJ1] add new operation BlackmanWindow 2022-06-03 11:19:50 +08:00
David 229603b4b5 Added docs to ut part 4 2022-06-02 21:37:15 -04:00
i-robot dccfbdc832
!34610 TruncateDiv and TruncateMod CPU ops
Merge pull request !34610 from jaspreetsinghsambee/truncate_ops_dev
2022-06-02 15:35:28 +00:00
i-robot 254e6f67c7
!35372 [MSLITE][CPU] avx512 hardward self feel function, remove nnacl/CMakeLists.txt avx512f flag and add avx512 hardward Inspection mechanism
Merge pull request !35372 from Greatpan/avx512_sf_remove_avx512f
2022-06-02 14:31:24 +00:00
i-robot 39799bad1b
!35295 Release nccl in release funtion
Merge pull request !35295 from ZPaC/finalize-nccl-in-release
2022-06-02 12:38:17 +00:00