Tianlei Wu
8a6f6bc38b
add --enable_cuda_line_info to build.py ( #6773 )
2021-02-22 22:00:21 -08:00
Guoyu Wang
f5313cc4ce
[CoreML EP] Add options to enable CoreML EP only on hardware with Apple Neural Engine ( #6765 )
2021-02-22 18:55:27 -08:00
Edward Chen
ee35be0129
Support specifying globally allowed types from build script ( #6677 )
...
Add initial support for constraining operator kernel implementations (which support this type-granularity) to a set of allowed types from scripts.
2021-02-22 14:05:00 -08:00
Ivan Stojiljkovic
c91f314217
Add robust dependency check for Python package ( #6436 )
...
* Add robust dependency check for Python package
* Add version_info.py to .gitignore
* Fix Linux build
* Fix Windows CPU build
* Fix Windows 32-bit build
* Minor tweak
* Generate version_info.py earlier in onnxruntime_python.cmake
* Print a user-friendly message if cuDNN is not found in
* Relax version requirements for CUDA 11 - only the major version has to match
* Fix PATH environment variable to include CUDA 11 in 'Python packaging pipeline' (Windows/GPU)
* Fix the build with cuDNN 7
2021-02-21 15:11:28 -08:00
Scott McKay
9f7dffb109
Make it easier to fold MakeString usages to reduce binary size. ( #6754 )
...
* Decay char[n] to char* so that we get better re-use of MakeString invocations.
2021-02-20 20:41:19 +10:00
Tianlei Wu
3bda7f4d36
Fix longformer parity and perf regression ( #6760 )
...
* add fast kernel back, update benchmark and conversion scripts
2021-02-19 21:47:36 -08:00
Sergii Dymchenko
58f3aca95d
Support keyword arguments for ORTModule ( #6539 )
...
* Support keyword arguments for ORTModule.
* Add backward workaround to the test.
* Specify test name directly without -k.
* Handle unused inputs removed by ONNX exporter.
2021-02-19 13:40:44 -08:00
Guoyu Wang
47519623cd
Fix iOS/macOS build warning for inconsistent symbol visibility settings ( #6750 )
...
* Fix ios/macos build warning
* Corrent a typo
* Fix test failure
* Fix custom op test failure
2021-02-19 12:19:32 -08:00
Nat Kershaw (MSFT)
c170061998
Removed BUILD.md from master as source now lives in gh-pages ( #6709 )
2021-02-19 11:34:21 -08:00
Faith Xu
20d6613efb
Add direct link to build instructions on readme ( #6729 )
2021-02-19 10:56:50 -08:00
Yulong Wang
a25c8e52a9
fix link to samples in nodejs README ( #6746 )
2021-02-19 10:56:22 -08:00
Brian Martin
67e7a9c02d
Merged PR 5703823: GitHub RI OnnxRuntime to DmlDev
...
https://microsoft.visualstudio.com/WindowsAI/_build/results?buildId=31602694&view=results
Related work items: #31680703
2021-02-19 17:37:31 +00:00
liqunfu
79b966b01a
. ( #6751 )
...
make ort training pert test green. remove unneeded yaml
2021-02-19 09:03:58 -08:00
Brian Martin
609d0ec953
Merge remote-tracking branch 'upstream/master' into dmldev_temp
2021-02-19 07:53:00 -08:00
M. Zeeshan Siddiqui
1a2f1bd23a
Enable external CUDA allocator in ORTModule. ( #6745 )
...
* Enable external CUDA allocator in ORTModule.
* Fix assert after unification of allocators.
* Update no grad memory test.
* update comments.
* fix provider options array when not sharing allocator.
2021-02-18 20:01:13 -08:00
ytaous
39d182f7fc
ORTModule - FastGeluFusion/fp16 fix and minor LayerNormFusion cleanup ( #6734 )
...
* fastgelu fix
* assert cast outputs
Co-authored-by: Ethan Tao <ettao@OrtTrainingDev4.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
2021-02-18 13:17:16 -08:00
Thiago Crepaldi
fb3f1f5cc1
Enable custom ops on ORTModule ( #6740 )
2021-02-18 09:08:10 -08:00
Chi Lo
67c478ede4
Entropy method for calibration-based quantization ( #6619 )
...
* Add entropy method
* Update pre/post-preprocessing of yolov3
* Code refactor
* Code refactor
2021-02-18 05:50:59 -08:00
S. Manohar Karlapalem
3722dd2692
Update docs MCR image publishing policy change ( #6743 )
...
ONNX Runtime 1.7 will the last release that will publish MCR
container images for ONNX Runtime with OpenVINO EP. From ONNX
Runtime 1.8 onwards, this will be discontinued. Users are advised
to switch to using PyPi packages or build their own containers
using dockerfiles.
2021-02-18 05:45:46 -08:00
stevenlix
53eb948f4c
Upgrade TensorRT to v7.2.2 ( #6452 )
...
* upgrade to TensorRT 7.2.2
* extend GPU tensorrt CI timeout to 150 minutes
* update docker image name
* disable user interaction to avoid tensorrt container stuck when install tzdata
* upgrade to libssl1.1 for ubuntu20.04
* remove libicu60 from ubuntu20.04
* add libicu66 for ubuntu20.04
* debug
* llvm
* llvm
* disable ReverseSequenceTest.InvalidInput
* disable ReverseSequenceTest.InvalidInput
* fix issues
* fix issues
* Update linux-gpu-tensorrt-ci-pipeline.yml
* disable warning 4458 for TensorRT parser
* update onnx-tensorrt submodule
* disable warnings for TensorRT parser
* update onnx-tensorrt submodule to include latest bug fixes
* update setup_env_trt
* update pool for win trt ci pipeline'
Co-authored-by: George Wu <jywu@microsoft.com>
2021-02-18 04:30:47 -08:00
Sherlock
b7b5612159
Merge pull request #6742 from microsoft/mzs/sync-from-master
...
Sync from master
2021-02-18 00:12:38 -08:00
Yufeng Li
b1a12b49b7
Avoid removing constant weight that is graph output ( #6735 )
2021-02-17 19:55:19 -08:00
Olivia Jain
ea3aee4d5f
Bumping up version to 1.7 ( #6736 )
...
* bumping up version to 1.7
* Windows AI should align with ORT Version
2021-02-17 19:07:38 -08:00
M. Zeeshan Siddiqui
40dda452cf
Merge branch 'master' of https://github.com/microsoft/onnxruntime into mzs/sync-from-master
2021-02-18 03:03:01 +00:00
M. Zeeshan Siddiqui
e44ac6524f
Plug n Allocate with external CUDA allocator via PyBind. ( #6679 )
2021-02-17 18:59:38 -08:00
liqunfu
dd8ef4409a
Liqun/migrate perf test ( #6733 )
...
move ort training perf tests to azure devops
2021-02-17 17:48:47 -08:00
liqunfu
2c5e603bad
Liqun/nuphar nuget ( #6656 )
...
create nuphar nuget with correct name
2021-02-17 16:13:07 -08:00
Thiago Crepaldi
21f9e32c60
Merge pull request #6714 from microsoft/thiagofc/merge-from-master
...
Merge master into thiagofc/ortmodule-api
2021-02-17 16:08:53 -08:00
Ramakrishnan Sivakumar
a5bef6886b
Threading support for Hybrid core architecture ( #6728 )
2021-02-17 15:35:07 -08:00
Guoyu Wang
6810d98ea3
Update links to gh-pages for ORT minimal documents ( #6721 )
...
* Fix broken link in ort minimal docs
* Update link of build.md to gh-pages
2021-02-17 14:34:50 -08:00
Justin Stoecker
af4e5c0c6e
Minor WinML model test skip name change
2021-02-17 14:27:58 -08:00
Maajid khan
b41e9b5d4c
[OpenVINO-EP] Fixes OpenVINO-EP build on windows ( #6726 )
...
* Fixes OpenVINO-EP windows build
Openvino EP build is broken on windows. The issue
is wchar_t is UTF-16 on windows while on other platforms
such as Linux and MacOS, wchar_t is UTF-32.
so wide Unicode string has to be converted to an UTF8 string
for sure on windows.
This commit fixes this issue.
2021-02-17 13:49:03 -08:00
Thiago Crepaldi
9d4b730e46
Fix merge leftover
2021-02-17 11:58:06 -08:00
M. Zeeshan Siddiqui
9853ef84f8
Reduce binary size, limit asynchronous/backgroud thread stuff to training only.
2021-02-17 11:51:09 -08:00
M. Zeeshan Siddiqui
5b7e7aaa45
Move event_pool and message_queue to core.
2021-02-17 11:50:56 -08:00
M. Zeeshan Siddiqui
eecce31a8b
Fix build, cleanup.
2021-02-17 11:50:41 -08:00
Thiago Crepaldi
3184c47ad1
Merge branch 'master' into thiagofc/merge-from-master
2021-02-17 11:49:52 -08:00
Brian Martin
14c9095d11
Merged PR 5694922: Onnxruntime GitHub to DmlDev RI
...
https://microsoft.visualstudio.com/WindowsAI/_build/results?buildId=31524603&view=results
Related work items: #31680703
2021-02-17 19:15:52 +00:00
Brian Martin
c31d278826
Merge remote-tracking branch 'upstream/master' into dmldev_temp
2021-02-17 11:01:56 -08:00
Yulong Wang
9a9202a218
[Node.js binding] update dependency typedoc ( #6720 )
2021-02-17 10:22:05 -08:00
Changming Sun
0be5475de6
Update packaging pipelines( #6664 )
2021-02-17 09:53:36 -08:00
Changming Sun
46c06f6ac7
Change Windows GPU CI pipeline to CUDA11 ( #6616 )
2021-02-17 09:44:44 -08:00
Changming Sun
eefeacd828
Skip running gpt2 model in C# x86 ( #6722 )
2021-02-17 09:37:16 -08:00
Derek Murray
b8d5fa812c
Fix typo in README.md ( #6713 )
...
Fixes #6710 .
2021-02-17 09:29:30 -08:00
Wei-Sheng Chin
9e67b88c83
Use local rank as GPU ID ( #6719 )
2021-02-17 22:42:54 +08:00
RandySheriffH
9043df8b66
Deprecate OMP from nuget pipeline (release:1.7) ( #6671 )
...
* deprecate OMP from nuget
* remove omp build
* remove
* add openmp build
* add variants
* rename package
* move GPU to no omp pipeline
* reset path
* switch to abs path
* reset path
* add cpu package
* remove obsolete name
* set package name
Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2021-02-17 00:03:44 -08:00
Suffian Khan
105883f4b8
remove longformer_global_impl.cu from hipify ( #6716 )
2021-02-16 22:26:18 -08:00
Hariharan Seshadri
aa2622efb2
Support multiple dynamic inputs in custom ops ( #6666 )
2021-02-16 20:54:30 -08:00
baijumeswani
01dfa8e125
Support non tuple return values from torch.nn.module ( #6660 )
...
* Support dictionary, namedtuples and huffingface ModelOutput type for model return values
2021-02-16 20:48:32 -08:00
Scott McKay
02c7873b0e
Update ORT model conversion script to support custom ops ( #6701 )
...
* Add support for custom ops library to the ORT model conversion script
Simplify model conversion now that we read ops from the ORT format model.
Enable custom ops in the python bindings if custom ops are turned on in a minimal build.
* Add test of model conversion involving custom ops.
2021-02-17 12:52:39 +10:00
