Commit Graph

5915 Commits

Author SHA1 Message Date
yanghaoran 8c444c00fc
回退 'Pull Request !36155 : [assistant][ResizeNearestNeighborV2][ResizeNearestNeighborV2Grad] cast shape type size_t to int64_t for ResizeNearestNeighborV2 & ResizeNearestNeighborV2Grad' 2022-06-22 01:22:24 +00:00
i-robot c39a0b5969
!36239 Enable syntax of "for x in xs", when xs is an interpret node
Merge pull request !36239 from LiangZhibo/for
2022-06-22 01:08:52 +00:00
i-robot 2089629480
!36129 Refactor fallback buildt-in function
Merge pull request !36129 from LiangZhibo/master
2022-06-22 01:08:42 +00:00
i-robot 3b035ebd9d
!36237 [MD] RandomResizedCrop - handle Interpolation and image input type mismatch
Merge pull request !36237 from cathwong/ckw_xtranuni_code9_randomreizedcrop
2022-06-21 15:32:55 +00:00
i-robot d1d447cfba
!35724 hcom subgraph reuse; send/recv distribute reorder
Merge pull request !35724 from lyqlola/hcom
2022-06-21 14:04:59 +00:00
i-robot e9ab472b31
!36232 Add ut for emebedding cache
Merge pull request !36232 from zyli2020/master
2022-06-21 13:35:14 +00:00
i-robot ac319b870d
!36155 [assistant][ResizeNearestNeighborV2][ResizeNearestNeighborV2Grad] cast shape type size_t to int64_t for ResizeNearestNeighborV2 & ResizeNearestNeighborV2Grad
Merge pull request !36155 from 李文轼/ResizeV2
2022-06-21 12:43:12 +00:00
i-robot dcd438e430
!33522 [assistant][ops] Add new aicpu operator SparseApplyAdagradDA
Merge pull request !33522 from 杨旭华/SparseApplyAdagradDA
2022-06-21 12:23:29 +00:00
liangzhibo 5a4d86caa8 for x in xs syntax when xs is numpy array 2022-06-21 15:09:35 +08:00
lizhenyu c536447125 add ut for emebedding cache 2022-06-21 11:55:03 +08:00
i-robot 976e13eff8
!36200 Fix PadToSize after Decode(to_pil=True)
Merge pull request !36200 from xiaotianci/fix_pad_to_size
2022-06-21 02:56:45 +00:00
i-robot ca563a9c56
!36236 [MD] MapDataset - correct count of old transforms
Merge pull request !36236 from cathwong/ckw_xtranuni_code10_mixed_list
2022-06-21 02:55:19 +00:00
liangzhibo 5eeba07b2d Refactor fallback builtin function 2022-06-21 10:17:03 +08:00
i-robot f208f1f24a
!36144 [Fallback] Check with the processing of interpret nodes during ast tree parsing.
Merge pull request !36144 from Margaret_wangrui/parse
2022-06-21 02:05:08 +00:00
i-robot 95a9cfa729
!35345 [AutoParallel] Optimize the Memory Usage for the CrossEntroyLoss
Merge pull request !35345 from huangxinjing/fix_cross_loss
2022-06-21 01:45:41 +00:00
Cathy Wong b77e8b9f08 [MD] RandomResizedCrop - handle Interpolation and image input type
mismatch
2022-06-20 20:41:42 -04:00
Cathy Wong 1230a962b9 [MD] MapDataset - correct count of old transforms 2022-06-20 16:12:53 -04:00
i-robot 0167368eb0
!36142 [MD] ToTensorOp: Add error msg for unsupported input image types: uint32, int64, uint64, string.
Merge pull request !36142 from cathwong/ckw_xtranuni_code8_totensor_more
2022-06-20 16:19:24 +00:00
i-robot 2919e625e3
!35699 Fix MultiHeadAttention dtype Error
Merge pull request !35699 from huangxinjing/fix_attention_dtype
2022-06-20 11:19:32 +00:00
i-robot 3568de5110
!32453 [assistant][ops] Add new Log2
Merge pull request !32453 from wxy220/log2
2022-06-20 10:00:05 +00:00
Margaret_wangrui 3570fef872 [Fallback] Check with the processing of interpret nodes during ast tree parsing. 2022-06-20 17:59:06 +08:00
i-robot 36698223e7
!36035 fix profiler missing aicpu and flops data bug,profiler init add aclrtSetDevice
Merge pull request !36035 from zangqx/zangqx_30
2022-06-20 09:50:13 +00:00
i-robot dccb95fc63
!34690 [assistant][ops] Add Real, Imag, Complex, Angle
Merge pull request !34690 from 李航/rica
2022-06-20 09:27:18 +00:00
i-robot acd22a8da0
!36049 [Fallback] Support print value which the node class type is numpy array.
Merge pull request !36049 from Margaret_wangrui/init_numpy
2022-06-20 08:20:33 +00:00
Xiao Tianci bf7fc7064d fix PadToSize with Decode 2022-06-20 15:41:36 +08:00
i-robot b07a681f4e
!35742 Delete redundant JSON calculation in the Select Kernel
Merge pull request !35742 from 王禹程/aicore
2022-06-20 05:53:35 +00:00
huangxinjing d0f2bf37c6 Add cross entory loss
[AutoParallel]Add sub graph support for parallel training
Fix forward mark error and virtual dataset error
Add Test cases
Add virtual dataset inserts, forbiden the subgraphs as candidates
2022-06-20 10:53:21 +08:00
Margaret_wangrui 17d5c0c2a1 [Fallback] Support class type is numpy array. 2022-06-20 09:56:27 +08:00
宋梦雨 2e71e26352 add the log2 2022-06-19 16:49:05 +08:00
Megalomania e84abd7bc7 [feat][assistant][I48O8C] add new aicpu operator SparseApplyAdagradDA 2022-06-19 12:36:51 +08:00
fujianzhao 862752fdce feat: add ops.applyAdadelta CPU backend, add nn.Adadelta, ops.applyAdadelta CPU backend support vmap. 2022-06-19 08:01:14 +08:00
i-robot e5c2dcbabb
!33167 [feat][assistant][I48O8Z] aicpu add new operator log
Merge pull request !33167 from 郑鹏飞/log
2022-06-18 14:12:37 +00:00
i-robot 4b4e53ec5a
!36025 [MD] decode string array in cpp when creating numpy from tensor - eager mode
Merge pull request !36025 from Mohammad Motallebi/text_decode_cpp
2022-06-18 11:38:02 +00:00
liyuanshuo123@yeah.net 0fb06e5c3e [feat][assistant][I4CRJO][I4CRJP] cast shape type size_t to int64_t for ResizeNearestNeighborV2 & ResizeNearestNeighborV2Grad 2022-06-18 17:40:09 +08:00
yanghaoran e2ee993d88
回退 'Pull Request !33227 : [assistant][ops] Add SegmentMean, SegmentProd' 2022-06-18 08:57:36 +00:00
yanghaoran 3193636d97
回退 'Pull Request !34402 : [assistant][ResizeNearestNeighborV2][ResizeNearestNeighborV2Grad] add new ascend operator ResizeNearestNeighborV2 & ResizeNearestNeighborV2Grad' 2022-06-18 08:14:42 +00:00
i-robot df3a288a89
!35256 shape function refactor
Merge pull request !35256 from chengbin/shape_func_refactor_v2
2022-06-18 07:03:10 +00:00
i-robot 1317769328
!34402 [assistant][ResizeNearestNeighborV2][ResizeNearestNeighborV2Grad] add new ascend operator ResizeNearestNeighborV2 & ResizeNearestNeighborV2Grad
Merge pull request !34402 from 李文轼/ResizeV2
2022-06-18 03:56:03 +00:00
i-robot 5bd1964bf7
!33227 [assistant][ops] Add SegmentMean, SegmentProd
Merge pull request !33227 from 王乐/SegmentMean
2022-06-18 03:43:43 +00:00
i-robot cf09e3582e
!35963 Fix HWC2CHW with 2 dim tensor
Merge pull request !35963 from xiaotianci/fix_hwc
2022-06-18 03:17:47 +00:00
i-robot 2dc90d35f4
!32927 [assistant][ops] Add DivNoNan
Merge pull request !32927 from 张渝/DivNoNan
2022-06-18 02:56:32 +00:00
zheng_pengfei 7fd8fe6ce7 [feat][assistant][I48O8Z] aicpu add new operator log 2022-06-18 09:06:02 +08:00
Cathy Wong 427fb21057 [MD] ToTensorOp: Add error msg for unsupported input image types: uint32, int64, uint64, string. 2022-06-17 17:17:31 -04:00
i-robot 8af9b70f47
!36030 [MD] ToTensor, TypeCast and ToType fixes, plus UT enhancements
Merge pull request !36030 from cathwong/ckw_xtranuni_code7_to_tensor_float16
2022-06-17 16:28:47 +00:00
ckey_Dou 0f4901b3ed shape function refactor: 2022-06-17 23:36:41 +08:00
i-robot f7aa696067
!36046 Support single interpret node in while condition
Merge pull request !36046 from huanghui/fix-fallback
2022-06-17 12:04:17 +00:00
liyiqi bed5e50ee6 hcom subgraph reuse; send/recv distribute reorder 2022-06-17 16:25:13 +08:00
i-robot e902c7c731
!35839 fix bug of moe
Merge pull request !35839 from bichaoyang/master
2022-06-17 08:15:25 +00:00
臧庆香 313095aac9 profiler init add aclrtSetDevice 2022-06-17 14:45:06 +08:00
lh735291378 0c714b9fc8 修改bprop
修改bprop

还原bprop

修改gradmathops

modify testops

delet 快捷方式

还原bprop

代码合入更改

修改冲突

还原bprop

修改 testops

修改 testops

修改 格式

修改格式

修改格式

修改grad

modify

修改格式

.c: return kFloat32 -> std::make_shared<TensorType>(kFloat32)

修改grad

修改格式

 改宏定义长度z

修改编译错误

add real imag complex angle
2022-06-17 13:06:06 +08:00