Merge from master to ort_training #3607
edgchen1
merged
49 commits into
ort_training
from
edgchen1/merge_from_master
Fix static analysis warnings found by VC++ (#3530)
7c89f38a
[Java] Add support for map and sequence information on output nodes (…
c9152723
Acl improvements (#3463)
93b957a5
Disable downloading test data on Windows (#3551)
1a222b3f
Dualapipartitionattibute.h header is missing in nuget package (#3350)
951484ba
perf tuning docs update (#3520)
92269ae4
hotfix for skiplayernorm (#3543)
fcb27c4e
Fork the WinML APIs into the Microsoft namespace (#3503)
2717c178
Update dependencies graph
209b41a6
update onnxruntime version string for telemetry (#3526)
43c3a5ed
Fix build break in mlas\lib\quantize.cpp: missing nearbyintf (#3572)
c7b6fab2
Make De/QuantizeLinear support half (#3531)
f822a548
remove option --enable_lto (#3515)
5acd8dbe
Fix `gen_doc` build option and refresh documentation (#3545)
b4457ecb
Update Attention operator for GPT2 (#3474)
5d3b2170
Add GPT2 Attention Fusion in optimization script (#3488)
7f46f347
Disabled failed maxpool test on GPU (#3549)
6fe688c7
Fix some typos. (#3582)
3e884b4b
Disable downloading test data on Linux (#3581)
d6824585
Fix some too popular warnings. (#3578)
38a18023
Implement Inverse(12) for CPU and CUDA (#3485)
db9566f7
Refactor some for the calibate.py. Add QLinearAdd and QLinearMul supp…
c3658228
Fix BatchNorm CUDA kernel definition
15995620
Implement OneHot CUDA Kernels (#3390)
56b223bc
Fix WinML namespace build break (#3583)
14e387aa
Generalize reshape fusion (#3554)
fcf0f6ee
Clean up build.py code (#3466)
ea62b343
Add ability to batch device copy for graph inputs and outputs. (#3580)
7d5348f8
Support conv transpos 1D in cuda provider. (#3300)
422266c4
Fix build break
75426a30
Added support to build onnxruntime with ACL (#3586)
381fee47
Fork WinML IDL Guids (#3591)
31b6629e
Remove openmp from gpu build
911d1253
Convert Gelu to use TryParallelFor (#3599)
2579a72a
Refactor - ScatterElements (#3559)
e233e6ba
Merge remote-tracking branch 'origin/master' into edgchen1/merge_from…
daa14b64
Fix merge issue.
87fad09c
debug option for dumping tensorrt subgraphs. (#3604)
1c37d5e6
[Java] Adding model metadata support (#3573)
3dd3f841
Threadpool related changes. (#3564)
9636da39
Fix GraphTransformationTests tests.
d50c3e7a
Revert "Convert Gelu to use TryParallelFor (#3599)"
28a0c863
Disable CudaKernelTest.SoftmaxCrossEntropyLoss_LargeSizeTensor becaus…
2e4b9b1d
Add --enable_onnx_tests to Windows builds to allow set up of test dat…
297ab43b
Add --skip_onnx_tests to orttraining Windows builds.
47f1758f
Improve TopK performance. (#3612)
b4508dbd
Refine BERT optimization script options (#3618)
d69bc313
pengwa
dismissed these changes
on 2020-04-22
Merge remote-tracking branch 'origin/ort_training' into edgchen1/merg…
8d09cefa
Merge remote-tracking branch 'origin/master' into edgchen1/merge_from…
8df5076d
SherlockNoMad
approved these changes
on 2020-04-22
Reviewers
Assignees
No one assigned
Labels
training
Milestone
No milestone
Login to write a write a comment.
Login via GitHub