Commit graph

6910 commits

Author SHA1 Message Date
Dmitri Smirnov
607b7df060
Allow saving on CPU usage for infrequent inference requests by reducing thread spinning (#11841)
Introduce Start/Stop threadpool spinning switch
Add a session config option to force spinning stop at the end of the Run()
2022-06-23 10:04:37 -07:00
pengwa
c398ad513f
Fix orttraining-linux-ci-pipeline - Symbolic shape infer (#11965)
fix symbolic shape error due to upgraded numpy + legacy sympy
2022-06-23 08:23:36 -07:00
Ye Wang
e24349b8f2
Optimize t5 encoder in beam search (#11926)
* ooptimize t5 encoder

* update

* update

* update

* refactor expand impl

* cuda tests passed

* update

* alignment

* more alignments

* review comments
2022-06-22 12:45:02 -07:00
Dwayne Robinson
f6d2fe8311
MeanVarianceNormalization CPU EP axes attribute validation (#11925)
Validate axes attribute parameter properly rather than silently returning incorrect results
2022-06-22 12:03:13 -07:00
Preetha Veeramalai
f54476a42f
Dll version fix ovep4.1 (#11953)
* Setting default version values for ovep dlls as well

* Update backend_manager.cc

Co-authored-by: mayavijx <mayax.vijayan@intel.com>
Co-authored-by: mohsin <mohsinx.mohammad@intel.com>
2022-06-22 11:09:36 -07:00
pengwa
2229c48547
fix mpi in training build (#11855)
fix mpi build for gcc8 or higher
2022-06-22 10:04:44 +08:00
Vincent Wang
03beed0ceb
Remove Cast before and after Gelu (#11885)
* fuse cast gelu

* use PropagateCastOps

* fix ut
2022-06-22 09:07:48 +08:00
Gary Miguel
4bf22e2a40
Update ONNX to 1.12 (#11924)
Follow-ups that need to happen after this and before the next ORT release:
* Support SequenceMap with https://github.com/microsoft/onnxruntime/pull/11731
* Support signal ops with https://github.com/microsoft/onnxruntime/pull/11778

Follow-ups that need to happen after this but don't necessarily need to happen before the release:
* Implement LayerNormalization kernel for opset version 17: https://github.com/microsoft/onnxruntime/issues/11916

Fixes #11640
2022-06-21 17:19:52 -07:00
Dwayne Robinson
64f95d400a
Update DML 1.9 Nuget package to fix WindowsAI nuget pipeline build issue (#11934) 2022-06-21 15:55:51 -07:00
Scott McKay
3b1224dc08
Add .net6 support to the C# nuget package. (#11908)
* Add .net6 support to the C# nuget package.

Currently requires jumping through a lot of hoops due to .net 6 only being supported in the preview release of VS 2022.

Build existing targets using msbuild.
Add .net6 targets and build using dotnet.
Create nuget package with combined targets.

A few misc automated changes from VS to spacing and adding a couple of properties.
2022-06-22 08:08:24 +10:00
Arseny
8c8a781cdb
fix: handle setBindingDimensions return value in TensorRT EP (#11929) 2022-06-21 14:30:27 -07:00
Edward Chen
5646410f65
Enable Pad test cases with initializer inputs only when building NNAPI EP on Android. (#11932) 2022-06-21 14:16:55 -07:00
sfatimar
61a74f2f4d
Mohsin/enable dynamic shapes (#11867)
* Add pypi build changes to latest Master

* Add ORT training part of OV build

* Disabling SqueezeOpTest.BadAxes

* Add ONNXruntime branch ARG to Docker build

* Changes to include file details versions

* Commit File Version Updates

* Change naming for linux build

* Add fix for pylint format errors

* Fix pylint warnings.

* Enable Dynamic Shapes for OV_API_20

* Update requirements.txt whl version- internal_ci fix

* Update backend_manager.cc MYRIAD Fix

* Update wheel version in requirements.txt

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update setup.py

* Fix pylint warnings

* Fix pylint warnings 2

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

* Update backend_manager.cc

Co-authored-by: Preetha Veeramalai <preetha.veeramalai@intel.com>
Co-authored-by: mayavijx <mayax.vijayan@intel.com>
Co-authored-by: Sahar Fatima <sfatima.3001@gmail.com>
Co-authored-by: mohsinmx <mohsinx.mohammad@intel.com>
2022-06-21 08:03:58 -07:00
Adrian Lizarraga
b20daeda81
Update Linux Multi GPU TensorRT pipeline to TensorRT 8.4 (#11923)
* Try manually installing trt8.4 in multi-gpu pipeline

* Remove stmts that clean up cmake, ctest. Update tensorrt repository name passed to get_docker_image.py

* Update trt and cudnn home

* Don't install trtexec cli tool.

* Increase job timeout

* Revert timeout change and use trt placeholder builder build option
2022-06-21 07:59:11 -07:00
Ye Wang
859ef277a0
apply zcode changes to the beam search op (#11880)
* apply zcode  changes to the beam search op

* fix pipeline failure

* add doc

* workaround for C#

* update

* update

* use name zcode

* review comment

* review comments

* fix cpplint

* review coments
2022-06-20 18:39:07 -07:00
RandySheriffH
cefceff5c9
Mark the end of APIs for release 1.12 (#11914)
* mark the end of APIs for 1.12

* add static assert for C API 1.12
2022-06-20 15:22:55 -07:00
Adrian Lizarraga
ca35ea417a
[EP-Perf] Install new wheel>=0.35.1 dependency (#11917) 2022-06-20 15:09:27 -07:00
Yi Zhang
7f1e9e8c67
Bash: there should be a whitespace after not operator. (#11910)
add whitespace after not
2022-06-21 05:14:32 +08:00
Chi Lo
457ce6cb89
Make symbolic shape inference script support external weight (#11909)
* add support for external data

* fix format

* fix format

* fix typo

* fix typo
2022-06-20 13:07:45 -07:00
Dwayne Robinson
c1577d08ca
DML EP QuantizeLinear defer axis validation for test_quantizelinear_cpu (#11906)
DML EP QuantizeLinear defer axis validation
2022-06-20 11:03:32 -07:00
Chih-Hsuan Yen
0736c604c7
Fix building DNNL EP with GCC 12 (#11667)
This issue is similar to [1], where template two-phase lookups break the
build. Apparently GCC 12 is now as strict as Clang:

```
[ 69%] Building CXX object CMakeFiles/onnxruntime_providers_dnnl.dir/build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc.o
In file included from /build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/common.h:36,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/shared_library/provider_api.h:19,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_subgraph.h:11,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.h:5,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc:3:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h: In instantiation of ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&) [with T = gsl::span<long int>; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   recursively required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = long unsigned int; Args = {const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = const char*; Args = {long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:46:17:   required from ‘std::string onnxruntime::detail::MakeStringImpl(const Args& ...) [with Args = {const char*, long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:93:32:   required from ‘std::string onnxruntime::MakeString(const Args& ...) [with Args = {char [20], long unsigned int, char [23], long int, char [9], gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc:276:11:   required from here
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: error: no match for ‘operator<<’ (operand types are ‘std::ostringstream’ {aka ‘std::__cxx11::basic_ostringstream<char>’} and ‘const gsl::span<long int>’)
   33 |   ss << t;
      |   ~~~^~~~
In file included from /usr/include/c++/12.1.0/istream:39,
                 from /usr/include/c++/12.1.0/sstream:38,
                 from /build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/common.h:27:
/usr/include/c++/12.1.0/ostream:108:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(__ostream_type& (*)(__ostream_type&)) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  108 |       operator<<(__ostream_type& (*__pf)(__ostream_type&))
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:108:36: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘std::basic_ostream<char>::__ostream_type& (*)(std::basic_ostream<char>::__ostream_type&)’ {aka ‘std::basic_ostream<char>& (*)(std::basic_ostream<char>&)’}
  108 |       operator<<(__ostream_type& (*__pf)(__ostream_type&))
      |                  ~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/ostream:117:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(__ios_type& (*)(__ios_type&)) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>; __ios_type = std::basic_ios<char>]’
  117 |       operator<<(__ios_type& (*__pf)(__ios_type&))
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:117:32: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘std::basic_ostream<char>::__ios_type& (*)(std::basic_ostream<char>::__ios_type&)’ {aka ‘std::basic_ios<char>& (*)(std::basic_ios<char>&)’}
  117 |       operator<<(__ios_type& (*__pf)(__ios_type&))
      |                  ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/ostream:127:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(std::ios_base& (*)(std::ios_base&)) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  127 |       operator<<(ios_base& (*__pf) (ios_base&))
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:127:30: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘std::ios_base& (*)(std::ios_base&)’
  127 |       operator<<(ios_base& (*__pf) (ios_base&))
      |                  ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/ostream:166:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(long int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  166 |       operator<<(long __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:166:23: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘long int’
  166 |       operator<<(long __n)
      |                  ~~~~~^~~
/usr/include/c++/12.1.0/ostream:170:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(long unsigned int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  170 |       operator<<(unsigned long __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:170:32: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘long unsigned int’
  170 |       operator<<(unsigned long __n)
      |                  ~~~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:174:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(bool) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  174 |       operator<<(bool __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:174:23: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘bool’
  174 |       operator<<(bool __n)
      |                  ~~~~~^~~
In file included from /usr/include/c++/12.1.0/ostream:833:
/usr/include/c++/12.1.0/bits/ostream.tcc:91:5: note: candidate: ‘std::basic_ostream<_CharT, _Traits>& std::basic_ostream<_CharT, _Traits>::operator<<(short int) [with _CharT = char; _Traits = std::char_traits<char>]’
   91 |     basic_ostream<_CharT, _Traits>::
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:92:22: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘short int’
   92 |     operator<<(short __n)
      |                ~~~~~~^~~
/usr/include/c++/12.1.0/ostream:181:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(short unsigned int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  181 |       operator<<(unsigned short __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:181:33: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘short unsigned int’
  181 |       operator<<(unsigned short __n)
      |                  ~~~~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/bits/ostream.tcc:105:5: note: candidate: ‘std::basic_ostream<_CharT, _Traits>& std::basic_ostream<_CharT, _Traits>::operator<<(int) [with _CharT = char; _Traits = std::char_traits<char>]’
  105 |     basic_ostream<_CharT, _Traits>::
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:106:20: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘int’
  106 |     operator<<(int __n)
      |                ~~~~^~~
/usr/include/c++/12.1.0/ostream:192:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(unsigned int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  192 |       operator<<(unsigned int __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:192:31: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘unsigned int’
  192 |       operator<<(unsigned int __n)
      |                  ~~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:201:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(long long int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  201 |       operator<<(long long __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:201:28: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘long long int’
  201 |       operator<<(long long __n)
      |                  ~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:205:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(long long unsigned int) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  205 |       operator<<(unsigned long long __n)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:205:37: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘long long unsigned int’
  205 |       operator<<(unsigned long long __n)
      |                  ~~~~~~~~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:220:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(double) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  220 |       operator<<(double __f)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:220:25: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘double’
  220 |       operator<<(double __f)
      |                  ~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:224:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(float) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  224 |       operator<<(float __f)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:224:24: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘float’
  224 |       operator<<(float __f)
      |                  ~~~~~~^~~
/usr/include/c++/12.1.0/ostream:232:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(long double) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  232 |       operator<<(long double __f)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:232:30: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘long double’
  232 |       operator<<(long double __f)
      |                  ~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:245:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(const void*) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>]’
  245 |       operator<<(const void* __p)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:245:30: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘const void*’
  245 |       operator<<(const void* __p)
      |                  ~~~~~~~~~~~~^~~
/usr/include/c++/12.1.0/ostream:250:7: note: candidate: ‘std::basic_ostream<_CharT, _Traits>::__ostream_type& std::basic_ostream<_CharT, _Traits>::operator<<(std::nullptr_t) [with _CharT = char; _Traits = std::char_traits<char>; __ostream_type = std::basic_ostream<char>; std::nullptr_t = std::nullptr_t]’
  250 |       operator<<(nullptr_t)
      |       ^~~~~~~~
/usr/include/c++/12.1.0/ostream:250:18: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘std::nullptr_t’
  250 |       operator<<(nullptr_t)
      |                  ^~~~~~~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:119:5: note: candidate: ‘std::basic_ostream<_CharT, _Traits>& std::basic_ostream<_CharT, _Traits>::operator<<(__streambuf_type*) [with _CharT = char; _Traits = std::char_traits<char>; __streambuf_type = std::basic_streambuf<char>]’
  119 |     basic_ostream<_CharT, _Traits>::
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:120:34: note:   no known conversion for argument 1 from ‘const gsl::span<long int>’ to ‘std::basic_ostream<char>::__streambuf_type*’ {aka ‘std::basic_streambuf<char>*’}
  120 |     operator<<(__streambuf_type* __sbin)
      |                ~~~~~~~~~~~~~~~~~~^~~~~~
In file included from /usr/include/c++/12.1.0/bits/basic_string.h:48,
                 from /usr/include/c++/12.1.0/string:53,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_subgraph.h:7:
/usr/include/c++/12.1.0/string_view:672:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, basic_string_view<_CharT, _Traits>)’
  672 |     operator<<(basic_ostream<_CharT, _Traits>& __os,
      |     ^~~~~~~~
/usr/include/c++/12.1.0/string_view:672:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   ‘gsl::span<long int>’ is not derived from ‘std::basic_string_view<_CharT, _Traits>’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/bits/basic_string.h:3883:5: note: candidate: ‘template<class _CharT, class _Traits, class _Alloc> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, const __cxx11::basic_string<_CharT, _Traits, _Allocator>&)’
 3883 |     operator<<(basic_ostream<_CharT, _Traits>& __os,
      |     ^~~~~~~~
/usr/include/c++/12.1.0/bits/basic_string.h:3883:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   ‘const gsl::span<long int>’ is not derived from ‘const std::__cxx11::basic_string<_CharT, _Traits, _Allocator>’
   33 |   ss << t;
      |   ~~~^~~~
In file included from /usr/include/c++/12.1.0/bits/ios_base.h:46,
                 from /usr/include/c++/12.1.0/streambuf:41,
                 from /usr/include/c++/12.1.0/bits/streambuf_iterator.h:35,
                 from /usr/include/c++/12.1.0/iterator:66,
                 from /usr/include/oneapi/dnnl/dnnl.hpp:28,
                 from /usr/include/dnnl.hpp:20,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_subgraph.h:10:
/usr/include/c++/12.1.0/system_error:279:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, const error_code&)’
  279 |     operator<<(basic_ostream<_CharT, _Traits>& __os, const error_code& __e)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/system_error:279:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘const std::error_code&’
   33 |   ss << t;
      |   ~~~^~~~
In file included from /usr/include/c++/12.1.0/memory:77,
                 from /usr/include/oneapi/dnnl/dnnl.hpp:29:
/usr/include/c++/12.1.0/bits/shared_ptr.h:70:5: note: candidate: ‘template<class _Ch, class _Tr, class _Tp, __gnu_cxx::_Lock_policy _Lp> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, const __shared_ptr<_Tp, _Lp>&)’
   70 |     operator<<(std::basic_ostream<_Ch, _Tr>& __os,
      |     ^~~~~~~~
/usr/include/c++/12.1.0/bits/shared_ptr.h:70:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   ‘const gsl::span<long int>’ is not derived from ‘const std::__shared_ptr<_Tp, _Lp>’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:507:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, _CharT)’
  507 |     operator<<(basic_ostream<_CharT, _Traits>& __out, _CharT __c)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:507:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   deduced conflicting types for parameter ‘_CharT’ (‘char’ and ‘gsl::span<long int>’)
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:517:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, char)’
  517 |     operator<<(basic_ostream<_CharT, _Traits>& __out, char __c)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:517:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘char’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:523:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, char)’
  523 |     operator<<(basic_ostream<char, _Traits>& __out, char __c)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:523:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘char’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:534:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, signed char)’
  534 |     operator<<(basic_ostream<char, _Traits>& __out, signed char __c)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:534:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘signed char’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:539:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, unsigned char)’
  539 |     operator<<(basic_ostream<char, _Traits>& __out, unsigned char __c)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:539:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘unsigned char’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:598:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, const _CharT*)’
  598 |     operator<<(basic_ostream<_CharT, _Traits>& __out, const _CharT* __s)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:598:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   mismatched types ‘const _CharT*’ and ‘gsl::span<long int>’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:302:5: note: candidate: ‘template<class _CharT, class _Traits> std::basic_ostream<_CharT, _Traits>& std::operator<<(basic_ostream<_CharT, _Traits>&, const char*)’
  302 |     operator<<(basic_ostream<_CharT, _Traits>& __out, const char* __s)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/bits/ostream.tcc:302:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘const char*’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:615:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, const char*)’
  615 |     operator<<(basic_ostream<char, _Traits>& __out, const char* __s)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:615:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘const char*’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:628:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, const signed char*)’
  628 |     operator<<(basic_ostream<char, _Traits>& __out, const signed char* __s)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:628:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘const signed char*’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:633:5: note: candidate: ‘template<class _Traits> std::basic_ostream<char, _Traits>& std::operator<<(basic_ostream<char, _Traits>&, const unsigned char*)’
  633 |     operator<<(basic_ostream<char, _Traits>& __out, const unsigned char* __s)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:633:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘t’ (type ‘const gsl::span<long int>’) to type ‘const unsigned char*’
   33 |   ss << t;
      |   ~~~^~~~
/usr/include/c++/12.1.0/ostream:754:5: note: candidate: ‘template<class _Ostream, class _Tp> _Ostream&& std::operator<<(_Ostream&&, const _Tp&)’
  754 |     operator<<(_Ostream&& __os, const _Tp& __x)
      |     ^~~~~~~~
/usr/include/c++/12.1.0/ostream:754:5: note:   template argument deduction/substitution failed:
/usr/include/c++/12.1.0/ostream: In substitution of ‘template<class _Ostream, class _Tp> _Ostream&& std::operator<<(_Ostream&&, const _Tp&) [with _Ostream = std::__cxx11::basic_ostringstream<char>&; _Tp = gsl::span<long int>]’:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   recursively required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = long unsigned int; Args = {const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = const char*; Args = {long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:46:17:   required from ‘std::string onnxruntime::detail::MakeStringImpl(const Args& ...) [with Args = {const char*, long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:93:32:   required from ‘std::string onnxruntime::MakeString(const Args& ...) [with Args = {char [20], long unsigned int, char [23], long int, char [9], gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc:276:11:   required from here
/usr/include/c++/12.1.0/ostream:754:5: error: no type named ‘type’ in ‘struct std::enable_if<false, void>’
In file included from /build/python-onnxruntime/src/onnxruntime/build/_deps/abseil_cpp-src/absl/container/inlined_vector.h:40,
                 from /build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/inlined_containers_fwd.h:18,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/shared_library/provider_api.h:21:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h: In instantiation of ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&) [with T = gsl::span<long int>; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   recursively required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = long unsigned int; Args = {const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:39:17:   required from ‘void onnxruntime::detail::MakeStringImpl(std::ostringstream&, const T&, const Args& ...) [with T = const char*; Args = {long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::ostringstream = std::__cxx11::basic_ostringstream<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:46:17:   required from ‘std::string onnxruntime::detail::MakeStringImpl(const Args& ...) [with Args = {const char*, long unsigned int, const char*, long int, const char*, gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:93:32:   required from ‘std::string onnxruntime::MakeString(const Args& ...) [with Args = {char [20], long unsigned int, char [23], long int, char [9], gsl::span<long int>}; std::string = std::__cxx11::basic_string<char>]’
/build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc:276:11:   required from here
/usr/include/c++/12.1.0/cstddef:123:5: note: candidate: ‘template<class _IntegerType> constexpr std::__byte_op_t<_IntegerType> std::operator<<(byte, _IntegerType)’
  123 |     operator<<(byte __b, _IntegerType __shift) noexcept
      |     ^~~~~~~~
/usr/include/c++/12.1.0/cstddef:123:5: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/include/onnxruntime/core/common/make_string.h:33:6: note:   cannot convert ‘ss’ (type ‘std::ostringstream’ {aka ‘std::__cxx11::basic_ostringstream<char>’}) to type ‘std::byte’
   33 |   ss << t;
      |   ~~~^~~~
In file included from /build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl:25,
                 from /build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/shared_library/provider_api.h:15:
/build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl-lite.hpp:1477:37: note: candidate: ‘template<class IntegerType, class> constexpr gsl::byte gsl::operator<<(byte, IntegerType)’
 1477 |   gsl_api inline gsl_constexpr byte operator<<(byte b, IntegerType shift) gsl_noexcept {
      |                                     ^~~~~~~~
/build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl-lite.hpp:1477:37: note:   template argument deduction/substitution failed:
/build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl-lite.hpp:237:5: error: no type named ‘type’ in ‘struct std::enable_if<false, gsl::detail::enabler>’
  237 |   , typename = typename std::enable_if<(VA), gsl::detail::enabler>::type
      |     ^~~~~~~~
/build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl-lite.hpp:1405:3: note: in expansion of macro ‘gsl_REQUIRES_T’
 1405 |   gsl_REQUIRES_T((std::is_integral<T>::value))
      |   ^~~~~~~~~~~~~~
/build/python-onnxruntime/src/onnxruntime/onnxruntime/gsl/gsl-lite.hpp:1476:31: note: in expansion of macro ‘gsl_ENABLE_IF_INTEGRAL_T’
 1476 |   template <class IntegerType gsl_ENABLE_IF_INTEGRAL_T(IntegerType)>
      |                               ^~~~~~~~~~~~~~~~~~~~~~~~
make[2]: *** [CMakeFiles/onnxruntime_providers_dnnl.dir/build.make:342: CMakeFiles/onnxruntime_providers_dnnl.dir/build/python-onnxruntime/src/onnxruntime/onnxruntime/core/providers/dnnl/subgraph/dnnl_reduce.cc.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:1611: CMakeFiles/onnxruntime_providers_dnnl.dir/all] Error 2
make: *** [Makefile:146: all] Error 2
```

Error messages above are from gcc 12.1.0 on Arch Linux.

[1] https://github.com/microsoft/onnxruntime/pull/10014
2022-06-20 10:41:04 -07:00
Dmitri Smirnov
267a424e52
Retry Rework execution frame to reduce memory allocations (#11897)
* Revert "Revert "Refactor ExecutionFrame and SessionState to reduce memory all… (#11888)"

This reverts commit d2cbae3a04.

* Revert prepacked_weights to avoid indirect inclusion in CUDA and TRT code that breaks the build.
2022-06-20 10:29:43 -07:00
Tianlei Wu
6ee2c1b5fc
Remove temperature input from BeamSearch operator (#11896)
* remove temperature input
* update index of remaining inputs
2022-06-20 09:50:45 -07:00
Chi Lo
eb41bfb7b5
Fix graph viewer to proto (#11862)
* Add test for case where main const initialier in subgraph

* update test to use trt ep

* add initializer when converting from graph viewer to proto

* add comments

* add comments

* add comments

* only add initialier that is outer scope value

* make including outer scope value optional

* modify python format

* modify python format

* modify python format

* Remove test

* remove redundant argument
2022-06-19 19:28:18 -07:00
sumitsays
52f2b3bf89
[DML EP] Remove suffix removal adhoc logic for fusedNodeArgNames (#11879)
* DML EP: Remove suffix removal hack for fusedNodeArgName

* Acknowledged PR comments

* Removed reference from gsl::span

Co-authored-by: Sumit Agarwal <sumitagarwal@microsoft.com>
2022-06-17 17:04:16 -07:00
sfatimar
f97bd38c4f
UEP 4.1 release (#11834)
* Add pypi build changes to latest Master

* Add ORT training part of OV build

* Disabling SqueezeOpTest.BadAxes

* Add ONNXruntime branch ARG to Docker build

* Changes to include file details versions

* Commit File Version Updates

* Change naming for linux build

* Add fix for pylint format errors

* Fix pylint warnings.

* Fix pylint errors - stage 2

Signed-off-by: Preetha Veeramalai <preetha.veeramalai@intel.com>

* Fix pylint errors - stage 3

* Fix pylint format - stage4

Signed-off-by: Preetha Veeramalai <preetha.veeramalai@intel.com>

* Commit for Wheel Release >0.35.1

Co-authored-by: Preetha Veeramalai <preetha.veeramalai@intel.com>
Co-authored-by: mayavijx <mayax.vijayan@intel.com>
Co-authored-by: Sahar Fatima <sfatima.3001@gmail.com>
Co-authored-by: nmaajidk <n.maajid.khan@intel.com>
2022-06-17 14:49:04 -07:00
Edward Chen
a93fe7824a
Update EP compile API deprecation warning message. (#11808)
Minor wording update to warning message to clarify that the function style Compile API is deprecated now and will be removed soon.
Also updated some code comments.
2022-06-17 12:49:24 -07:00
Yi Zhang
f70201c801
Make sure the command works in both centos and ubuntu. (#11894)
make one bash condition compatible with POSIX
2022-06-17 12:19:22 -07:00
Rachel Guo
1494120423
[NNAPI EP] Unsqueeze op support (#11864)
* wip

* save unsqueeze support

* minor update

* remove unnecessary line

* address pr comments

* add comments
2022-06-17 12:07:18 -07:00
Yi-Hong Lyu
4ac72e305c
NHWC Resize optimization (#11825)
The optimization consists of:

* Use int32_t instead of int64_t
* Use different code path for tf_crop_and_resize or other
  coordinate_transformation_mode to avoid redundant conditions
* Loop-invariant code motion of offset, coefficient and extrapolation_value
  check
* Use fixed point to avoid floating-point computation

Besides, it always transforms NCHW Resize to NHWC because it has higher perf in
the NHWC variant when the input X is 4D int8/uint8 tensor and the mode is
linear on ARM.

It improves DeepLab V3 with int8 quantization by 26%~27% on big core and 37% on
LITTLE core on AArch64. It also improves DeepLab V3 with uint8 quantization by
24%~25% on big core and 34% on LITTLE core on AArch64.

Co-authored-by: Yufeng Li liyufeng1987@gmail.com
2022-06-17 11:00:36 -07:00
Edward Chen
adcf7e66c8
[NNAPI EP] Pad Op (#11860)
Add basic support for Pad Op in NNAPI EP.
2022-06-17 10:05:31 -07:00
Adrian Lizarraga
ad4abbd75e
[EP-Perf-Dashboard] Add support for TensorRT 8.4 to EP Perf Dashboard (#11876)
Co-authored-by: George Wu <jywu@microsoft.com>
2022-06-17 09:16:51 -07:00
Yi Zhang
8bb0062873
add manylinux_2_27 CPU wheel (#11886)
* add manylinux_2_27

* minor refactory

* change base image

* minor refactor

* add tests

* fix condition
2022-06-17 19:38:38 +08:00
Yi Zhang
d2cbae3a04
Revert "Refactor ExecutionFrame and SessionState to reduce memory all… (#11888)
Revert "Refactor ExecutionFrame and SessionState to reduce memory allocations and improve data locality (#11804)"

This reverts commit 2ecba6fd25.
2022-06-17 17:07:21 +08:00
stevenlix
bd65acd08d
Share execution context memory between TensorRT subgraphs (#11859)
* share trt context memory

* update parser to 8.4-EA

* update parser to 8.4-GA

* add context memory sharing enable option

* update parser to 8.2-GA

* fix format issue

* reverse orders

* fix format

* fix format

* fix issues
2022-06-16 22:42:40 -07:00
Changming Sun
10478a09ca Revert "add manylinux_2_27 wheel (#11832)"
This reverts commit bbace23d0c.
2022-06-16 18:28:12 -07:00
Dmitri Smirnov
2ecba6fd25
Refactor ExecutionFrame and SessionState to reduce memory allocations and improve data locality (#11804)
Refactor ExecutionFrame and SessionState for better data locality and less memory allocations.
2022-06-16 16:50:48 -07:00
Dwayne Robinson
3d99f16e98
Merge pull request #11827 from microsoft/user/dwayner/DmlEp1.9
Integrate WindowsAI feature branch with DML EP features and DML 1.9
2022-06-16 13:04:00 -07:00
George Wu
df5ee6aa4e
[TensorRT EP] support TensorRT 8.4 (#11866)
* update trt 8.4ga

* trt 8.4 linux ci pipeline

* fix cmake

* placeholder_builder

* trt 8.4 windows pipeline

* gpu package pipeline

* trt 8.4.1.5 , packaging pipeline updates

* python packaging

* ctest timeout

* python packaging test

* bump timeout

* python format

* format

* revert

* newline

* enable trt python tests

* typo

* python format

* disable on windows
2022-06-16 07:46:40 -07:00
Dwayne Robinson
fe7b8b80ae Revert BatchNormalization change for now, falling back to CPU on mixed types until a more advanced solution is written 2022-06-15 21:49:18 -07:00
Dwayne Robinson
babd6e3fcd Update DirectML preview package with unmangled names 2022-06-15 18:16:58 -07:00
Maxiwell S. Garcia
3f8c9146d5
ppc64le: specialize generic 'mlas' functions to use VSX instructions (#11845) 2022-06-15 16:49:38 -07:00
Scott McKay
d64f23fec0
EP factory creation cleanup and enhancements. (#11798)
* Rework the EP factory creation setup so we're not cut-and-pasting function declarations in multiple places.
Convert append EP for SNPE to be generic, and also use for XNNPACK.
Add XNNPACK to C# API

* Don't need stub for MIGraphX as it's using provider bridge.

* Remove old 'create' functions that aren't applicable now that the EPs are built as separate libraries.

* Only use EPs that require the layout transform if the opset is supported by the layout transformer.

* Update wasm registration of xnnpack.
2022-06-16 07:01:41 +10:00
Rachel Guo
1a1c360a80
[NNAPI EP] Add Gather op support (#11824)
* initial gather support nnapi

* update

* minor update

* address pr comments

* add int32 indices test case for nnapi

* remove nnapi ep limitation for added UT

* add link for memcpy type punning usage
2022-06-15 09:44:07 -07:00
Vincent Wang
02457ec30a
[CUDA] GatherElements[Grad]/ScatterElements Bugfix and Perf Improve (#11374)
* gather elements bugfix and perf improve

* fix win build

* fix ut on some eps

* ut change

* resove comments

* resolve comments

* fix win build
2022-06-15 16:29:17 +08:00
Xavier Dupré
a805a49363
Move OrtValueVector from onnxruntime-training to onnxruntime (#11176)
* Move OrtValueVector from onnxruntime-training to onnxruntime

* disable dlpack on onnxruntime

* disable dlpack

* dlpack

* opaque inlcuded in any cc file of the python binding

* fix type issue

* fix incomplete name

* remove len()

* remove unused parameter

* black

* black

* black

* remove unused import

* add unit test to check the output type

* black

* lint

* lint

* lint

* fix method name

* Update onnxruntime/python/onnxruntime_pybind_ortvalue.cc

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/python/onnxruntime_pybind_ortvalue.cc

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/python/onnxruntime_pybind_ortvalue.cc

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/python/onnxruntime_pybind_ortvalue.cc

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/python/onnxruntime_pybind_ortvalue.cc

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/test/python/onnxruntime_test_python_sparse_matmul.py

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* Update onnxruntime/test/python/onnxruntime_test_python_sparse_matmul.py

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>

* check return type of C API

* lint

* lint

* fix missing ;

* fix type issue

* fix merge issue

Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>
2022-06-15 09:36:28 +02:00
Dwayne Robinson
ff8b173286 Typo in DirectML.Debug.dll 2022-06-15 00:18:40 -07:00
Dwayne Robinson
508c76a246 Add missing DirectML.Debug.dll 2022-06-15 00:16:10 -07:00
Dwayne Robinson
e3ec30efb6 Add missing GELU to ApiHelpers.h 2022-06-14 23:28:15 -07:00
Dwayne Robinson
4c1a410d54 Unmangle DML preview package filenames 2022-06-14 23:12:58 -07:00