8da4c1a763
synchronize latest ascend software 04 Jun 2020
6 years ago
f837ddc956
- Bug when empty strings sent to Python
...
- Support accepting Numpy of str as input
- Support batching strings
- Core logic of batch&pad is static
- Make Pad a utility function
6 years ago
b16a552d41
Revert "Revert "add pattern AdjustAllReduceMulAdduse the old opadd test case for bugtemp fix try""
...
This reverts commit 705c71a257
6 years ago
beb714d2d0
!1911 add a function to charge the node input and output is a scalar
...
Merge pull request !1911 from lianliguang/add-a-function-to-charge-the-node-input-or-output-if-is-a-scalar
6 years ago
2ff9e74d07
Add unique process for duplicated indices
6 years ago
9808e47663
change checkAicpu to CheckAICPU & add charge Scalar function to charge the input or output is scalar
6 years ago
b4c0ed4b36
add signle batchnorm fission pass
6 years ago
7465abc798
optimize transdata for pynative
6 years ago
1973594bd1
!1774 [MD] minddataset support padding samples
...
Merge pull request !1774 from liyong126/mindrecord_padded_sample
6 years ago
b236beae28
!1615 convert constant bool tensor to bool
...
Merge pull request !1615 from amongo/GetBoolValueFromConstantTensor
6 years ago
feff8899ac
support padding samples
6 years ago
c51d90d84e
!1767 Move LayerNormGrad split pass ahead of kernel select
...
Merge pull request !1767 from huanghui/LayerNormGrad-split-pass
6 years ago
769ae609b4
!1808 consistent design for num_samples
...
Merge pull request !1808 from Jamie/numsamples
6 years ago
beacc26077
* add isconstant primitive
...
* add infer_value for common math ops
* convert constant bool tensor to bool value
* do not infer value when encounter 0 as division
for while condition, do not unrool if condition is a tensor
6 years ago
e87ac6525e
Add batch norm fusion pattern for mix precision
6 years ago
cf87218fb7
place layernormgrad split pass before kernel select
6 years ago
51bc0c0460
consistent design for num_samples
...
updates
more update
more work
more fixin
post rebase updates
clang formatting
code review recovery
ci fixes
updates
update
update
update
6 years ago
0426ed057b
!1777 Remove ZerosLikeTensor and sub with ZerosTensor
...
Merge pull request !1777 from BowenK/master
6 years ago
96379faa3a
Remove ZerosLikeTensor and sub with ZerosLike
6 years ago
d1cec14a0c
add 2 pattern for softmaxgradext fusion pass
6 years ago
c63729b8e6
support mix target
6 years ago
cc9c004bc1
!1696 Enable ConfusionMulGrad fusion pass
...
Merge pull request !1696 from huanghui/TMP
6 years ago
692f38ded2
!1532 clean pylint warning in test dir
...
Merge pull request !1532 from jinyaohui/pylint1
6 years ago
59683a1d90
!1692 Fix topk bug for fasterrcnn
...
Merge pull request !1692 from meixiaowei/master
6 years ago
71acaa5300
enable ConfusionMulGrad fusion pass in bert only
6 years ago
387bcece9e
!1666 Add 5 patterns for AdamApplyOneWithDecay fusion pass
...
Merge pull request !1666 from huanghui/AdamApplyOneWithDecay-fusion-pass
6 years ago
86d197dfeb
clean pylint
6 years ago
ff05aa1faa
add newly 5 patterns for AdamApplyOneWithDecayRule fusion pass
6 years ago
e6f1cfa581
Check the input size of BatchNorm before fission in bert
6 years ago
1778ec0135
fix topk bug
6 years ago
99ca6b3e80
add SoftmaxGradExt fusion pass
6 years ago
8556eee3fc
!1578 recitify pretrained path and revert AdjustAllReduceMulAdduse
...
Merge pull request !1578 from gengdongjie/master
6 years ago
9c858444b0
!1568 Fix pylint warning for test cases of optimizer, infer, parse and pynative.
...
Merge pull request !1568 from seatea/fix-pylint-warning
6 years ago
04398cf88e
!1433 add tensor_minnie and separate py from ir
...
Merge pull request !1433 from leopz/tensor_base
6 years ago
705c71a257
Revert "add pattern AdjustAllReduceMulAdduse the old opadd test case for bugtemp fix try"
...
This reverts commit 3db8cfa54f
6 years ago
4508134ceb
add tensor_minnie and separate py from ir
6 years ago
c086d91aaf
!1505 Add some checks in ConstToAttr[StridedSliceGrad] pass
...
Merge pull request !1505 from huanghui/stride-slice
6 years ago
7398526abc
Fix pylint warning for test cases of optimizer, infer, parse and
...
pynative.
6 years ago
ee087bdf60
Check the size of topk input names before converting input to attr
6 years ago
3b80c10b1e
Fix some timing hole in these two testcases
6 years ago
e311473401
!1453 Raise mse threshold in RandomCropDecodeResizeOp
...
Merge pull request !1453 from Alexey_Shevlyakov/fix_random_crop_decode_resize_ut
6 years ago
1d65ae598a
extract const_to_attr_strided_slice_grad pass
6 years ago
5b9c145ff8
!1383 keep different attributes for cnode evaluation
...
Merge pull request !1383 from amongo/KeepPrimAttrInCNode
6 years ago
5e2f440eed
!1271 refactor of memreuse allocator to adapt the control stream
...
Merge pull request !1271 from yangjie159/refactor_memreuse_allocator
6 years ago
650a45b233
!1490 Fix pylint warning
...
Merge pull request !1490 from seatea/fix-pylint-warning
6 years ago
c6d8a4dc98
Fix pylint warning.
6 years ago
94818cf255
Fix fusion condition of transpose and reshape
6 years ago
cbf5390b34
refactor memreuse allocator
...
remove stream id from membuf
add kernel dependence to determine whether is usable
6 years ago
1159f2b11e
add prim infer attr cache
...
modify infer to get abstract value and attr value map together
support get attr in specialize
clone primitive when specialize if attribute is not the same with
infered attribute
6 years ago
2215e3267f
!1419 remove old buffer fusion pass
...
Merge pull request !1419 from Etone.Chan/Resnet50
6 years ago
a8fa847556
Allow join_all to be blocking
6 years ago
7a807cbff9
fix ut
6 years ago
42d724d8b4
remove old buffer fusion pass
6 years ago
e8980ed298
!1406 Simplify CondVar class
...
Merge pull request !1406 from JesseKLee/CondVar
6 years ago
a42dd21430
add 3 pattern for lamb_next_mv_rule fusion pass
6 years ago
848d19207f
!1359 Optimize the IR modules.
...
Merge pull request !1359 from ZhangQinghua/master
6 years ago
dbb86cb1be
Adjust some routines of FG and FGM, about the nodes info. IF.
6 years ago
f31564ce98
Remove the useless collectors in manager.
6 years ago
641112a40c
Initial Drop
6 years ago
daccfef738
!1361 Refactor multiple output pass
...
Merge pull request !1361 from huanghui/LambNextMVRule-fusion-pass
6 years ago
c5ac2cc38c
add memory copy module
6 years ago
f16ff539ba
refactor multiple patterns pass
6 years ago
964a757db2
!1378 Just want to debug this case in Jenkin environment
...
Merge pull request !1378 from JesseKLee/cond_var_debug
6 years ago
06ae724261
Again
6 years ago
458436186c
!1365 Clean up work for text python sub-package
...
Merge pull request !1365 from h.farahat/text_namespace
6 years ago
6f733ec113
!1308 Stage 2 of adding support for string Tensor
...
Merge pull request !1308 from h.farahat/string_tensor2
6 years ago
f967700ef4
!1215 [Auto parallel] Changing succive edges orders and add some strategy checkings
...
Merge pull request !1215 from Xiaoda/change-succive-edges-order-and-add-checking
6 years ago
3363d4e834
!1249 Add GNN dataset processing API
...
Merge pull request !1249 from heleiwang/hlw_gnn_data
6 years ago
6c21e556c4
Clean up work for text python package
6 years ago
4bdfddedb2
!1358 Link child graphs
...
Merge pull request !1358 from zhoufeng/link-child-graph-2
6 years ago
af5019b94f
link child graphs
...
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
6 years ago
599a449e0b
Support processing GNN data
6 years ago
fbdba6e4da
clean pylint
6 years ago
df361d1d26
Change mem layout of string tensor
...
add support for MindRecord and TFRecord
----
optimize tensorshape
optimize tensorshape and FlatIndex
TFRecord and MindRecord support for string tensor
Modify mem layout
Add new constructor
Add method Allocate
Change some GetMutableBuffer usages to AllocateBuffer
6 years ago
9f4b8a3cd1
changing the succive edges order in GetAliveSuccEdges() so that Triangle and Star Elimination can be merged into particular node; adding some check information
6 years ago
d9c74e0acd
!1328 Add 3 patterns support for LambNextMVWithDecay fusion pass
...
Merge pull request !1328 from huanghui/LambNextMvWithDecayRuleConds-fusion-pass
6 years ago
2e3d55ed87
!1281 Implementation of SplitOp
...
Merge pull request !1281 from Peilin/splitOp
6 years ago
71e8bb1960
general split case done, chaining sampler (basic case) is working
...
implementation 99% complete
everything and tested except for repeatable shuffling
tested most basic/typical split usecases
cleanup
some more cleanup
fix CI
more ci fix
more ci fixes
more ci fix
more ci fix
more ci fix
added more tests, fixed some bugs
some more clean up and test cases
added shard/shuffle before split warning/error
addressed code review comments and ci
fixed ci
6 years ago
07b14f32b9
!1296 move signature to primitivepy and bprop_func to utils
...
Merge pull request !1296 from leopz/fix_primitive
6 years ago
eaff850f11
add 3 patterns for lamb_next_mv_with_decay_rule pass
6 years ago
e4795974be
!1266 Refactor LambNextMVWithDecayRule fusion pass
...
Merge pull request !1266 from huanghui/LambNextMvWithDecayRuleConds-fusion-pass
6 years ago
5a68eba585
Refactor LambNextMVWithDecayRule fusion pass
6 years ago
451c20a6f5
Add UnicodeCharTokenizer for nlp
6 years ago
04763b8b76
move signature to primitivepy and bprop_func to utils
6 years ago
163b6b7ea7
add jieba c++ code
6 years ago
97ba539e7b
!1269 fix issue of loadding control input tensors failed in control sink mode
...
Merge pull request !1269 from wenchunjiang/fix_task_sink_bug
6 years ago
729ea8cc55
fix bug of loadding ctrl input tensors failed in control sink mode
6 years ago
04ac611fe8
!1189 Decoupling py default param from Parameter
...
Merge pull request !1189 from leopz/master
6 years ago
40e15996b0
move default_param out of parameter and remove pybind11 in anf define
6 years ago
3db8cfa54f
add pattern AdjustAllReduceMulAdduse the old opadd test case for bugtemp fix try
6 years ago
c793540cc9
!1231 Assign label resource for new control sink
...
Merge pull request !1231 from zhoufeng/label-assign
6 years ago
c99f904276
assign label resource for new control sink
...
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
6 years ago
828d0b124e
!1209 add format trans function
...
Merge pull request !1209 from liubuyu/master
6 years ago
1b7cdc4c61
!1218 clean pylint warning in test about import
...
Merge pull request !1218 from jinyaohui/pylint1
6 years ago
f70429d671
add format transfer
6 years ago
c8d33568f2
Add an new output to FusedMulApplyMomentum
6 years ago
5a914994ba
clean pylint
6 years ago
c4af71e236
add LarsV2 fission pass
6 years ago
bcfaff97f9
clean pylint
6 years ago
e8ca243364
-Add DE_STRING
...
-replace switch'case by indexing
- Add test case
- Add constructors
- Add getItem string
- Fix bugs
- Add more tests
- Tensor iterator
- asNumpy
- TextFileDataset
- Tensor(Numpy)
- Super > 2D
- Add more test cases for GeneratorDataset
- Change StartAddr to GetBuffer and GetMutableNuffer
- Raise an error if batch is used with strings
Clean-up work
6 years ago
ca64bfdbab
!1120 Code refactoring ：modify Infer to Eval
...
Merge pull request !1120 from Kang/master
6 years ago
889d7b6eea
!1180 [pre_activate]Fix pylint warning
...
Merge pull request !1180 from YuJianfeng/master
6 years ago
efb9672849
!1118 Add BatchNormGrad2BNInferGrad pass
...
Merge pull request !1118 from huanghui/batchnormgrad-to-bninfergrad-pass
6 years ago
c680cfbf27
!1157 dataset: add concat operation for dataset
...
Merge pull request !1157 from ms_yan/concat_dataset
6 years ago
ab031ee9ea
!1126 VOCDataset support object detection function
...
Merge pull request !1126 from xiefangqi/voc_support_detection
6 years ago
0f42fdbb80
Fix pylint warning
6 years ago
709828a98b
add BatchNormGrad2BNInferGrad pass
6 years ago
42ba885e58
!1098 Add BatchNorm2BNInfer pass
...
Merge pull request !1098 from huanghui/add-batchnorm-bninfer-pass
6 years ago
c937bad53f
minddata support voc
6 years ago
c0fa7b4b19
init commit of concat dataset
...
change to use __add__ operation instead ds.concat
6 years ago
be60bd3d79
Code refactoring for the static_analysis : modified Infer to Eval.
6 years ago
6db71e6576
add batchnorm2bninfer pass
6 years ago
2b061c84c7
Add batch_norm_grad infer fisson
6 years ago
597933b04b
!1113 rename kernel/mng to kernel/rts
...
Merge pull request !1113 from dengwentao/rts
6 years ago
6e89ebe6f0
Add BatchNorm fusion pattern with mix precision
6 years ago
4bb46606db
!1149 fix pylint in tests
...
Merge pull request !1149 from panyifeng/fix_pylint
6 years ago
755ba75deb
fix pylint
6 years ago
c4a5bfb787
add matmul eltwise buffer fusion pass
6 years ago
9e116e15fb
!1139 Change Thread::Join to async call with time out
...
Merge pull request !1139 from JesseKLee/timing_hole
6 years ago
1a052f7155
!1131 [Data]fix cmake build format
...
Merge pull request !1131 from xulei/filter_op
6 years ago
f2be49d85e
Fix a racing condition
6 years ago
903b64daa2
fix cmake build format
6 years ago
8b98f921cc
!1124 Add broadcast fusion pass
...
Merge pull request !1124 from YuJianfeng/master
6 years ago
fa83fc3e02
rename kernel/mng to kernel/rts
6 years ago
1a98c6b459
!999 [MD] mindrecord support reading file list
...
Merge pull request !999 from liyong126/mindrecord_file_list
6 years ago
31d8db9fc6
Add broadcast fusion pass
6 years ago
86d4797399
!1079 Convert AiCpu kernel when aicore not supported
...
Merge pull request !1079 from lianliguang/convert-to-AICPU-when-AiCore-not-supported-kernel
6 years ago
aa3f89e74f
mindrecord support read file list
6 years ago
691b0648e3
convert unsupported kernel in aicore to aicpu
6 years ago
85ff90c237
Add input shape condition for transpose_reshape fusion pass
6 years ago
3c643e7298
Move from dataset_dev branch to here
6 years ago
3f323e48e2
Add submodule id for log interface
6 years ago
f0bf438a55
reshape strategy search
6 years ago
9433ca6468
!998 format the func name
...
Merge pull request !998 from guozhijian/enhance_format_func_name
6 years ago
cf352d190f
format func name for mindrecord
6 years ago
12eaaf710b
mul_add_fusion pass supports when add's 2nd is mul
6 years ago
c2d364a573
batch with padding implemented
...
support for 1 specific dimension to be None, added validator
fix various CI complains
another round of CI fixes
ci
refactor parts of the code
code refactor
ci fix
comments added, fix bugs
address review comments
address review comments
review cmts
added simple perf test script
update pad code
perf imprv
6 years ago
def8573275
implementing-searching-strategy-for-inference
6 years ago
aa6f808616
Add batch norm bert fission pass
6 years ago
9a2daae84c
del unused change_mode_op
6 years ago
99c9b48a4d
delete TestHWBatchNormGradSplit
6 years ago
b323199dc1
syn code 430
6 years ago
7185961e89
Enable BatchNorm fusion pass
6 years ago
db3dd02838
!847 Optimize the collectors in manager which listen to the graphs and nodes changes.
...
Merge pull request !847 from ZhangQinghua/opt
6 years ago
2303453753
!869 Random data op
...
Merge pull request !869 from JesseKLee/random_data_op
6 years ago
732c7f8266
!794 fix issues image gradients 4d check
...
Merge pull request !794 from zhaozhenlong/fix-issues-image-gradients-complement-check
6 years ago
270bf831a9
Random Data Op
6 years ago
98939d839c
!856 Fix skip op bug
...
Merge pull request !856 from jiangzhiwen/dataset/skip_thread
6 years ago
66e7a36846
ImageGradients check 4d
...
fix DiagPart constant folding issue
fix argmin output type check
fix atan2 doc error
fix remove FusedBatchNorm and its grad
6 years ago
34bfa2f7c9
fix skip
6 years ago
yanghaoran
hesham
zhaoting
mindspore-ci-bot
yujianfeng
WilliamLian
huanghui
chujinjin
liyong
huangdongrun
Jamie Nisbet
mindspore-ci-bot
BowenK
kswang
jinyaohui
meixiaowei
gengdongjie
leopz
seatea
Jesse Lee
yangjie159
Alexey Shevlyakov
etone-chan
Zhang Qinghua
lizhenyu
zhoufeng
heleiwang
Xiaoda Zhang
Peilin Wang
qianlong
xulei2020
wenchunjiang
Wei Luning
liubuyu
mindspore-ci-bot
xiefangqi
ms_yan
Kang
panyifeng
dengwentao
fary86
yao_yf
jonyguo
Zirui Wu
changzherui
zhaozhenlong
jiangzhiwen