saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-02 03:55:34 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
1084 commits 2856 branches 57 tags 1.3 GiB
Commit graph

1084 commits

This branch
This branch
All branches
Author SHA1 Message Date
Hariharan Seshadri
4d768b3a0f
Fix inclusion of ARM binary in the release pkg (#1513) (#1521) 
* Fix inclusion of ARM binary in the release pkg

* Add lib and pdb as well
 2019-07-31 15:57:03 -07:00
shahasad
fb5d0fc538
Publish nuget package to azure blob store (#1525) 
Publish daily build NuGet package to Azure blob store for sharing among internal partners
 2019-07-31 14:17:54 -07:00
Tracy Sharpe
0b0e32909a
NCHWc: Enable Conv/Add fusion for stride=2 convolutions (#1518) 
Update the NCHWc graph transformer to allow Conv/Add fusion for convolutions where stride=2.
 2019-07-31 12:30:05 -07:00
Scott McKay
14d46ee890
Init prev_Ht for zero length sequence to avoid valgrind warning. (#1516) 
Couple of performance cleanups
  - don't create debug label string unless dumping matrixes
  - use raw pointer in fill_n calls
 2019-07-31 14:46:00 +10:00
Jorgen Thelin
fb7bdd177b Profiler-IsEnabled (#1503) 
Avoid use of Hungarian naming convention for cross-platform API code.

I'm taking my cue here from the "ONNX Runtime coding conventions and standard" document which say we use the "Google C++ style guide", and that says "Do not use Hungarian notation"
https://github.com/microsoft/onnxruntime/blob/master/docs/Coding_Conventions_and_Standards.md
https://google.github.io/styleguide/cppguide.html#Windows_Code

X-ref: internal PR 4824
 2019-07-30 13:32:01 -07:00
shahasad
a86486ab7f
Post binary sizes to dashboard database (#1517) 
Python script and necessary changes in the azure-pipelines yaml file to post the binary size data from NuGet package build. Currently only posted from CPU pipeline. GPU and other pipelines may be added as necessary.
 2019-07-30 08:59:43 -07:00
Pranav Sharma
44ab301586
More C API changes. (#1519) 
* Mention OrtCreateSessionFromArray in C API doc

* Cleanup a few inconsistencies in the C API.

* updates

* More updates
 2019-07-29 18:35:28 -07:00
Dwayne Robinson
cf73f63cb9 Enable float16 MatMul+Add -> GEMM fusion for performance boost (#1506) 2019-07-29 15:18:02 -07:00
Ke Zhang
cf5a4b5856
remove the GetStream from cuda ep. (#1514) 
* remove the GetStream from cuda ep.

* fix comments
 2019-07-29 15:01:29 -07:00
Yufeng Li
d6a30485be
Rename Tensor.Size() to Tensor.SizeInBytes() (#1502) 
Rename Tensor.Size() to Tensor.SizeInBytes()
 2019-07-26 14:15:53 -07:00
Hariharan Seshadri
6f538dc861
Support missing optional attribute in Squeeze operator (#1505) 
* Make Squeeze operator support no axes attribute cases

* Fix build break

* Resolve PR comments and exclude tensorrt for the new tests
 2019-07-26 11:16:35 -07:00
Hector Li
717e764e8e
Move Class CudnnDropout to cudnn_common.h (#1492) 
1. Move non_max_suppression_test.cc to object_detection folder

2. Move Class CudnnDropout to cudnn_common.h so that can share it with other ops. Move the cuda memory allocation part out of CudnnDropout to avoid memory leak.
 2019-07-26 10:41:13 -07:00
Emma Yu
8589be69b2 Organized build instructions (#1504) 2019-07-26 09:12:24 -07:00
xkszltl
33ae28ccb1 Empty double quota "" is passed to find_package(Thread), causing a test command gcc ... "" ... failed while trying to compile a source file with empty name. (#1508) 
```
[user@******** /]# gcc ""
gcc: error: : No such file or directory
gcc: fatal error: no input files
compilation terminated.
```
 2019-07-26 03:11:37 -07:00
xkszltl
be16b274fc Upgrade mklml and set march with official option. (#1469) 
1. There's formal way for setting march.
2. Upgrade to new MKLML.

Besides, the mem patch can be drop for v1.0.0 since it's fixed in upstream.
 2019-07-25 19:37:59 -07:00
Changming Sun
1a115ed8cb
Fix buffer overrun bug in CPU upsample op (#1501) 2019-07-25 17:05:05 -07:00
Hariharan Seshadri
6df4bc2ebe
Update scripts to access pipeline variables correctly (#1499) 
* Update scripts to access IsReleaseBuild pipeline variable correctly

* Correct access of PACKAGENAME pipeline variable

* Fix Linux CUDA 10 package tests

* Enable C# GPU test

* Update
 2019-07-25 15:30:32 -07:00
Changming Sun
e0829b2b13
Revert the last changes on tree ensemble classifier (#1498) 
Revert PR #1015 and PR #1276 . Because PR #1015 is causing test failures, but I can't revert it individually. I have to revert it together with #1276
 2019-07-25 13:55:10 -07:00
Changming Sun
4ace393bea Fix sign-compare warnings with gcc 2019-07-25 13:00:51 -07:00
Changming Sun
6625eecd09
Temp fix for a crash in fused graph (#1488) 2019-07-25 13:00:25 -07:00
Changming Sun
a7223ed801
Fix android build (#1489) 2019-07-25 13:00:00 -07:00
Scott McKay
f052966972
Remove special casing of "None" as a dim_param (#1482) 
* Remove special casing of "None" as a dim_param
 2019-07-25 17:18:14 +10:00
Yufeng Li
a8e3ff47fd
Add no scale check for resize and upsample (#1484) 
Description: Describe your changes.
Add no scale check for resize and upsample
Motivation and Context

Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
 2019-07-24 22:29:52 -07:00
shahasad
258ff06e42 Revert "publish nuget package to azure blob (#1309)" (#1485) 
This reverts commit 1601650161.
 2019-07-24 18:07:33 -07:00
daquexian
ec3c553501 NNAPI EP Update (#1483) 
* Update DNNLibrary

* Allow fp16 by default

* Add nnapi build in ci

* Fix nnapi ep after #1268

* Remove unused variables

* Support nnapi in onnx_test_runner

* Update DNNLibrary to fix tests

* Update build.py for android build support, solve conflict of
tools/ci_build/build.py

* Support non-ARM Android build, solve conflict of tools/ci_build/build.py

* Enable android test by x86_64 android emulator

* Add dnnlibrary/NNAPI support in build.py

* suppress the verbose adb output

* Remove debug logs

* Install cmake by pip

* Fix undefined host_protoc_path

* cmake==3.13.2 in pypi is actually 3.12.2, so install 3.13.2.post1 instead

* Fix Android ARM64 build

* Use android ndk r20 instead of r19c, fix conflicts in install_deps_android.sh
 2019-07-24 13:20:05 -07:00
avidiyal
c0f927c57c docker updated to support openvino R1.1 (#1475) 
* docker updated to support openvino R1.1

* Update README.md

Updated Readme to downlaod openvino R1.1
 2019-07-24 13:16:23 -07:00
Changming Sun
91d32c9060
Add docs for the fns candy demo (#1479) 2019-07-24 08:38:04 -07:00
jignparm
a8e9e1878e
Reduce artifacts size (#1477) 
* Update wildcard pattern to match only relevant archives

* Update TensorRT build to add CUDA VS extensions
 2019-07-23 22:23:51 -07:00
Yufeng Li
bb26865758
Optimize the resize and upsample (#1426) 
Description: Describe your changes.
Optimize the resize and upsample operators
Motivation and Context

Why is this change required? What problem does it solve?
For case with input with shape [1,128, 267, 200] and scales [1, 1, 1.97, 2], Resize and upsample get 15x gain (w/o: 1020ms, w: 71ms on my local box). It should benefit other scenarios at similar level.
If it fixes an open issue, please link to the issue here.
 2019-07-23 21:39:54 -07:00
Pranav Sharma
4aa4ca1502
Relax shape validation checks. Log a warning instead of returning an error. (#1476) 
* Mention OrtCreateSessionFromArray in C API doc

* Relax shape validation checks. Log a warning instead of returning an error.
 2019-07-23 18:47:46 -07:00
Hariharan Seshadri
c5f2f0f15b
Upgrade version number for ORT in preparation for release (#1468) 
* Update version number to 0.5.0 in preparation for release

* Update to README.md to direct to Versioning doc

* Resolve PR comment

* Remove incorrect line generation

* Minor updates to update version script

* Minor comment update
 2019-07-23 16:33:06 -07:00
shahasad
1601650161 publish nuget package to azure blob (#1309) 2019-07-23 11:07:35 -07:00
Changming Sun
be02214a17
Add a comment to onnxruntime_cxx_inline.h (#1466) 2019-07-23 08:45:37 -07:00
jignparm
b41f6eef52 Jignparm/copy cuda extensions (#1462) 
* Add CUDA extensions for v 10.0

* Add CUDA extensions for v 10.0

* update path

* change 'vsts' to 'github'
 2019-07-22 23:27:48 -07:00
Pranav Sharma
818c023535
Add/correct missing SAL annotations + avoid using unsigned types (except where counts are involved). (#1451) 
* Add/correct missing SAL annotations + other cosmetic changes.

* Add Outptr

* Don't use unsigned types
 2019-07-22 23:25:53 -07:00
Scott McKay
387d4c72bb
Strip invalid dim_param and dim_value values out. Allow re-use in event of shape mismatch if buffer is large enough (#1439) 
* Remove invalid dim_param and dim_value values when creating a NodeArg.
* Allow re-use of a large enough buffer if there's a shape mismatch.
* Update handling in python to treat unset dimension the same as a dim_param (equivalent to None).
* Fix GetTensorShapeFromTensorShapeProto to handle neither dim_param and dim_value being set.
 2019-07-23 14:55:54 +10:00
Jesse Benson
bbe92035c6 OStreamSink produces interleaved output due to issuing multiple operator<< calls to the underlying stream. (#1465) 2019-07-23 14:54:24 +10:00
shahasad
768ced703c
Expose provider factory C API, especially for CUDA users (#1461) 
Exposed provider factory C API, for cpu and cuda providers, into the published packages.
 2019-07-22 19:03:06 -07:00
Hector Li
31838fc9ee
remove const_cast which makes it's not thread safe. (#1463) 2019-07-22 17:55:29 -07:00
Yufeng Li
6be93f11e5
build mklml/ngraph without openmp (#1460) 
cleanup the option to build mklml/ngraph without openmp
 2019-07-22 16:59:32 -07:00
Faith Xu
1f13a9f982
Update to include more samples (#1381) 
* Update to include more samples

* Link fix
 2019-07-22 16:48:26 -07:00
Hariharan Seshadri
1fc6f8ee5b
Support double type for a few ops (#1450) 
* Initial commit

* More ops

* fix missing declarations for ReduceSum and ReduceSumSquare

* Add tests for new ops supporting double

* isable Add_dobule for OpenVINO EP
 2019-07-22 16:44:55 -07:00
Changming Sun
9d67292c8c
Document for the C/C++ samples (#1442) 
1. Document for the C/C++ samples.
2. Fix a null pointer errorin the imagenet sample.
 2019-07-22 16:14:49 -07:00
Pranav Sharma
29de25c5a7 Mention OrtCreateSessionFromArray in C API doc (#1459) 2019-07-22 15:44:46 -07:00
jywu-msft
a5f57f43c2
fix mkldnn linux build break. (#1458) 2019-07-22 12:22:49 -07:00
Hector Li
f938a6e53a
Add test for LSTM/GRU which has shorter sequence in the middle (#1437) 
Add test for LSTM/GRU which has shorter sequence in the middle
 2019-07-22 11:42:57 -07:00
Klein Hu
227734139a Fix ORTSRV nightly build (#1440) 
* Update the build_dir

* Fix indent in the model_zoo_tests.py

* Remove unnecessary tests in the server build.
 2019-07-21 19:12:05 -07:00
kile0
ac25a2643b add VS2019 CMake generator instrs (#1441) 2019-07-20 21:57:10 -07:00
jignparm
2c05291908
Jignparm/patch 0001 (#1419) 
* remove extra $ from 8592

* fix
 2019-07-20 17:07:56 -07:00
jignparm
1a957e0642
Update C-API packaging pipeline to use CUDa 10 (#1445) 2019-07-20 14:27:43 -07:00