Commit graph

537 commits

Author SHA1 Message Date
Edward Chen
0ebeaf529d
Check kernel def hashes (#7120)
Add unit test for verifying kernel def hashes.
Add way to add new types to kernel definition without changing hash.
2021-04-01 17:42:58 -07:00
sfatimar
52bcef4d4f
Openvino ep 2021.3 (#7180)
* Integrate openvino-ep-2021.3

* operators type

* changed the myriad as it is case sensitive

* logging information for openvino-ep-2021.3

* Unit test fix

* Resize operator added for myriad

* Fixed python tests for CPU and GPU

* data commit for loop tile and gatherelements failure

* adding checks for Where

* fixing gatherelements and loop tests

* disabling instance normalization test for now as there seems to be a
myriad bug, putting loop in ops supported only because all the tests
fail

* gather elements op test taking care of warning message

* condition needs to be an intializers

* Disabled python test for Myriad

* Disable compilation warning for MSVC windows compiler

* softmax_test, threedimaxis0 and 1 test give accuracy mismatch
tensoroptest disables test gives accuracy mismatch
gather test gives accuracy mismatch

* Updated with ov version 2021.3

* Updated with ov version 2021.3

* Updated README

* Disabling python tests for cpu

* Disabling python tests with accuracy mismatch on cpu

* Added fix for Linux CI Pipeline failure

-> Disabled tests that were throwing segfault

Co-authored-by: sfatimar <sahar.fatima@intel/com>
Co-authored-by: MaajidKhan <n.maajidkhan@gmail.com>
Co-authored-by: Aravind <aravindx.gunda@intel.com>
2021-04-01 11:28:54 -07:00
Edward Chen
04679e31ab
Specify CUDA compute capability 7.5 in Linux GPU build (#7203)
Recently a build agent pool was changed to use T4 GPUs (CUDA compute capability 7.5). Updating some CUDA build options to accommodate that.
2021-03-31 18:51:44 -07:00
liqunfu
e545604499
. (#7165) 2021-03-30 13:58:30 -07:00
Changming Sun
bbcf419ac6
Move the Windows GPU machine pool of Onnxruntime packaging pipelines to a new one (#7161) 2021-03-29 17:32:03 -07:00
Suffian Khan
f27835c4de
Disable batch size test for AMD CI pipeline after agent upgrade to Rocm 4.1 (#7153)
* disable batch size test for rocm 4.1 until resolved

* Update orttraining-pai-ci-pipeline.yml

Forgot to modify both pipelines
2021-03-26 22:32:39 -05:00
Suffian Khan
2b31b80b1f
icnrease timeout (#7145) 2021-03-26 11:26:18 -07:00
Changming Sun
2e3bbad19f
Move TensorRT Windows CI build to the machine pool (#7127) 2021-03-24 14:28:25 -07:00
baijumeswani
a7a2a16edd
Pass arguments to azure_scale_set_vm_mount_test_data from perf test ci pipeline (#7094) 2021-03-22 21:48:32 -07:00
liqunfu
309885b08d
upload ort-gpu-training python nightly package to azure feed (#6998) 2021-03-22 18:44:54 -07:00
Thiago Crepaldi
867804bea1
Add auto doc gen for ORTModule API during CI build (#7046)
In addition to ORTModule auto documentation during packaging, this PR also update golden numbers to fix CI
2021-03-22 10:20:33 -07:00
Edward Chen
8d5bfdeb47
Increase timeout for Android CI pipeline by 30 minutes. (#7065) 2021-03-19 08:03:22 -07:00
Changming Sun
701e73b5b8
Move Linux minimal build CI pipeline to the new Linux machine pool (#7050) 2021-03-18 12:09:12 -07:00
Xavier Dupré
514444d820
Fix pipeline generating python documentation (#7027)
Co-authored-by: xavier dupré <xavier.dupre@gmail.com>
2021-03-17 16:57:51 -07:00
Thiago Crepaldi
335edaa2c4
Merge pull request #6973 from microsoft/thiagofc/merge-ortmodule-into-master
Introduce ORTModule training API to ONNX Runtime
2021-03-17 10:30:06 -07:00
Changming Sun
ed2d441a2e
Update ORT server build pipeline (#7030)
1. Migrated it to Ed's new docker build script
2. Use python 3.6 instead, because it is the default one in ubuntu 18.04
3. Move the "pip install" command to the docker image build stage(instead of when running the image)
2021-03-16 18:02:09 -07:00
Changming Sun
2361cb99b6
Remove CentOS CI pipeline (#6997) 2021-03-16 10:55:03 -07:00
Tiago Koji Castro Shibata
975e4efb8a
Package ARM artifacts (#6805) 2021-03-16 10:12:48 -07:00
Changming Sun
4161758058
Remove openmp related packaging pipeline (#6991)
1. Remove openmp related packaging pipelines and build jobs.
2. Set continueOnError to true for the TSAUpload tasks. Their service is unstable recently.
3. Update Ubuntu 16 docker images to Ubuntu 18, in prepare for getting C++17 support
4. Cherry-pick the changes in 1.7.1 to the master: updating CFLAGS/CXXFLAGS to strip out debug symbols
2021-03-12 10:02:59 -08:00
Xavier Dupré
694389a85d
Automate generation of python documentation (#6909)
Co-authored-by: xavier dupré <xavier.dupre@gmail.com>
2021-03-11 19:02:45 -08:00
Thiago Crepaldi
89d450697b Introduce ORTModule training API to ONNX Runtime 2021-03-10 10:48:10 -08:00
Edward Chen
b6c4a7ac54
Support required types when excluding typed registrations (#6871) 2021-03-08 08:22:07 -08:00
RandySheriffH
f986ffcb5f
move pipeline file and change relative path (#6882)
Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2021-03-04 15:31:42 -08:00
Tiago Koji Castro Shibata
fa8d1b44b8
Fix app packaging in UWP (#6804)
* Change msbuild condition for UAP

* update .netcore target as well

* create nuget packages with _native path

* validate path under _native directory for windowsai package

* pep8

* add diagnostic error message

* pep8

* use baseame

* lib\uap10.0

* uap10

* build\\uap10.0

* Manually binplace winmds into appx when PackageReference is used.

* always binplace winmd regardless of packagereference since c# should work with packages.config also

* resolve all paths to full paths to avoid some reference warnings

* move winmds out of lib folder to prevent automatic component registration

Co-authored-by: Sheil Kumar <sheilk@microsoft.com>
2021-03-04 11:16:25 -08:00
Baiju Meswani
d5667554e6 Merge branch 'master' of github.com:microsoft/onnxruntime into bmeswani/merge_master_onto_ortmodule 2021-03-03 20:37:29 -08:00
Guoyu Wang
36a44d55ed
Only report Android Baseline binary size for master branch (#6844)
* Only report binary size from master

* update script

* Correct the typo
2021-03-01 15:57:18 -08:00
M. Zeeshan Siddiqui
ca48310d6d Merge branch 'master' of https://github.com/microsoft/onnxruntime into mzs/ortmodule-api-sync-from-master-210226 2021-02-27 04:25:23 +00:00
baijumeswani
c1b0cf6d0b
Add pipeline to clear the cache for huggingface transormers models (#6813) 2021-02-26 10:39:22 -08:00
baijumeswani
fa8a9015bd
Mount hf model cache and use cache for loading hf models (#6810) 2021-02-25 13:30:14 -08:00
Olivia Jain
db05d53b94
Setup perf in docker and add features (#6582)
* setup scripts to run in docker 
* percent threshold for accuracy 
* branch testing
2021-02-25 09:31:03 -08:00
Suffian Khan
8a148e44fb
make ci pipeline also run batch and convergence test (#6798) 2021-02-24 20:18:03 -08:00
baijumeswani
65ba51d93e
Re-enable test and increase timeout (#6785) 2021-02-23 18:51:06 -08:00
Edward Chen
5db0c9c648
Enable CI to cover globally allowed types (#6778)
Add test to CI build to cover type reduction with globally allowed types.
2021-02-23 10:24:12 -08:00
liqunfu
79b966b01a
. (#6751)
make ort training pert test green. remove unneeded yaml
2021-02-19 09:03:58 -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
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
liqunfu
dd8ef4409a
Liqun/migrate perf test (#6733)
move ort training perf tests to azure devops
2021-02-17 17:48:47 -08:00
Thiago Crepaldi
3184c47ad1 Merge branch 'master' into thiagofc/merge-from-master 2021-02-17 11:49:52 -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
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
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
RandySheriffH
c36ee4bd40
Rename Python packaging pipelines (#6682)
* rename pipelines

* resync and rename

* resync master

* rename package id

* remove OrtPackageId which is for nuget

Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2021-02-16 11:43:03 -08:00
RandySheriffH
497eef8d3d
remove omp (#6675)
Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2021-02-16 11:42:32 -08:00
Changming Sun
d48a4c0a54
Add CG step to nuget GPU pipeline (#6678) 2021-02-16 09:46:20 -08:00
Changming Sun
9a01174037
Disable some unit tests for training (#6699) 2021-02-16 09:45:59 -08:00
RandySheriffH
df3d6bad5f
Deprecate OMP from Python package (#6610)
1. For previous openmp build, remove --use_openmp, so thread pool will become default;
2. For previous non-openmp build, add --use_openmp and rename the package to indicate the inclusion.
3. Add a mac build with openmp enabled.
2021-02-12 21:50:41 -08:00
Scott McKay
25f7c93504
Require explicit inclusion of custom op support in a minimal build (#6663)
* Remove support from custom ops from the base minimal build as they contribute too much binary growth to an Android build.
Add ability to explicitly enable custom op support in a minimal build.
Change one minimal build CI to test adding custom op support (unit tests are run in that build to validate)
2021-02-13 12:42:33 +10:00
Changming Sun
dd50c39ac6
Change Linux python packaging pipeline compile flags (#6668) 2021-02-12 15:28:56 -08:00
Sheil Kumar
87cb6fd495
Add LearningModelBuilder to WinML Experimental Namespace along with various Audio operators (#6623)
* model building

* fix build

* winml adapter model building api

* model building

* make build

* make build again

* add model building with audio op

* inplace and inorder fft

* add ifft

* works!

* cleanup

* add comments

* switch to iterative rather than recursive and use parallelization

* batched parallelization

* fft->dft

* cleanup

* window functions

* add melweightmatrix op

* updates to make spectrogram test work

* push latest

* add onesided

* cleanup

* Clean up building apis and fix mel

* cleanup

* cleanup

* naive stft

* fix test output

* middle c complete

* 3 tones

* cleanup

* signal def new line

* Add save functionality

* Perf improvements, 10x improvement

* cleanup

* use bitreverse lookup table for performance

* implement constant initializers for tensors

* small changes

* add matmul tests

* merge issues

* support add attribute

* add tests for double data type windowfunctions and minor cleanup

* stft onesided/and not tests

* cleanup

* cleanup

* clean up

* cleanup

* remove threading attribute

* forward declare orttypeinfo

* warnings

* fwd declare

* fix warnings

* 1 more warning

* remove saving to e drive...

* cleanup and fix stft test

* add opset picker

* small additions

* add onnxruntime tests

* add signed/unsigned

* fix warning

* fix warning

* finish onnxruntime tests

* make windows namespace build succeed

* add experimental flag

* add experimental api into nuget package

* add experimental api build flag and add to windows ai nuget package

* turn experimental for tests

* add minimum opset version to new experimental domain

* api cleanup

* disable ms experimental ops test when --ms_experimental is not enabled

* add macro behind flag

* remove unused x

* pr feedback

Co-authored-by: Sheil Kumar <sheilk@microsoft.com>
2021-02-12 14:17:10 -08:00