Commit graph

8097 commits

Author SHA1 Message Date
Scott McKay
549cbc7e69
Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)
### Description
<!-- Describe your changes. -->
Fix issue with schema lookup where there are custom ops using the ONNX
domain.

Update testing infrastructure to use an explicit domain for custom ops.
Using an empty string clashes with the ONNX domain and can cause
unexpected issues. It's also a bad example for external users as our
docs point to the unit tests.

Fix a couple of places using exact matching of the node since version to
be slightly more flexible and use a range (which aligns with how the
kernel lookup works).

### 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. -->
Fixes a problem that came up when adding support for standalone custom
ops in an ORT format model. Separating these changes out to simplify
review.
2023-02-03 08:05:18 +10:00
Yulong Wang
cfb6e528c8
[js/web] remove 'module' field from package.json (#14532)
### Description
this is a workaround for
[#14529](https://github.com/microsoft/onnxruntime/issues/14504) when
consuming onnxruntime-web as ES module.
2023-02-02 13:46:57 -08:00
Justin Stoecker
03cfb7d73e
Use ORT node names in DML graphs/ops (#14461)
### Description
Applies ORT node names to corresponding compiled operators or DML graph
nodes.

### Motivation and Context
This makes it easier to correlate ONNX nodes to events in PIX GPU
captures when using the DML EP. Names set in the DML graph nodes require
additional modifications to the DML runtime library (available in a
future NuGet package).
2023-02-02 13:42:15 -08:00
Xavier Dupré
0bcca7ad45
Fix Gather to Split optimizer (#14478)
### Description
Gather to Split optimizer fails if opset == 18. This PR fixes one bug
and extend unit tests.



### Motivation and Context
The model produced by the optimizer does not follow onnx specifications
with opset 18.
2023-02-02 13:29:44 -08:00
Baiju Meswani
3d8fa4d77b
GetTrainingApi to not print to stderr when not an ort training build (#14515) 2023-02-02 13:28:32 -08:00
Baiju Meswani
68a402e739
Add support for python 3.10 for onnxruntime-training cuda and cpu (#14100) 2023-02-02 11:32:41 -08:00
RandySheriffH
01cafe89f0
Specify deps in deps.txt and manifest (#14530)
Specify new deps and update cgmanifest.json.

---------

Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2023-02-02 09:44:57 -08:00
pengwa
62442c3d27
Enable multiple step run for adamw tests (on device training) (#14520)
(cherry picked from commit 414b73a02123b672e496326664cd2dc3bd6c6d24)

### Rework for PR https://github.com/microsoft/onnxruntime/pull/14068:
Enable multiple step run for adamw tests (on device training)
### Removed duplicated MACRO checks for training.


### 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. -->
2023-02-02 18:40:30 +08:00
Baiju Meswani
7954976e0a
Fix python packaging pipeline (#14533)
fix onnx and protobuf inconsistencies in python packaging pipeline.
2023-02-02 13:11:18 +08:00
Yulong Wang
0578eeff91
upgrade EsrpCodeSigning from v1 to v2 (#14531)
### Description
This change upgrade EsrpCodeSigning from v1 to v2 in our build pipeline.
2023-02-02 13:08:26 +08:00
Dmitri Smirnov
61e7636e61
Re-work GetAvailableProviders API (#14486)
### Description
Re-work `OrtApi::GetAvailableProviders` in a way that the data is
returned in a single allocation.
Fix exception safety issues and fix `Release` function. 
Remove warning suppressions.
Fix exception safety issue in C++ API.
Fix exception safety issue in C# API.
Move EP name length enforcement to the implementation.

### Motivation and Context
The original motivation comes from
https://github.com/microsoft/onnxruntime/issues/14378.
However, the API is already implemented.

Cc: @prabhat00155
2023-02-01 14:38:04 -08:00
Yufeng Li
d9e675a2af
Do not fuse DQ+Node+Q if DQ produces graph output (#14509)
Fix issue #14501
2023-02-01 13:36:47 -08:00
Abhishek Jindal
3d388a1aea
change deepspeed version in warning from 0.7.3 to 0.8.0 (#14527)
### Description
change deepspeed version in warning from 0.7.3 to 0.8.0



### Motivation and Context
The version was updated for Deepspeed support in ORT from 0.7.3 to 0.8.0
but wasn't updated in the warnings message and this PR is to fix that.
2023-02-01 12:00:43 -08:00
Yulong Wang
b1a17188a6
[js/web] add LRN unpacked kernel for webgl backend (#14459)
### Description
add LRN unpacked kernel for webgl backend
2023-02-01 11:51:10 -08:00
Ye Wang
bc8686d365
Fix prefast warning 11263 (#14508)
### Description
<!-- Describe your changes. -->
https://dev.azure.com/aiinfra/ONNX%20Runtime/_workitems/edit/11263/


### 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. -->
2023-02-01 10:45:48 -08:00
Numfor Tiapo
3cc81460e0
Register ScatterElements-16 (#14425)
This PR registers ScatterElements-16 to the DML EP
- CPU fallback is added if the reduction attribute is in use, as this is
not yet supported by DML.

---------

Co-authored-by: Numfor Mbiziwo-Tiapo <numform@microsoft.com>
2023-02-01 09:46:37 -08:00
pengwa
a1ed703b73
Fix Python format CI failures due to black && pyright package updates (#14521)
### Fix failures due to black && pyright package updates

#### Problem

In the passing 3 hours, all PRs have their Python lint CI failed. Many
files are reported not well formatted. I believe this is due to updated
black package did some changes. Also there is a pyright checking
failure, after investigation it is due to pyright package upgrade.

##### Failure 1: "Lint Python" failure related to pyright: 
```
Run jordemort/action-pyright@v1
Run $GITHUB_ACTION_PATH/script.sh
🐶 Installing reviewdog ... https://github.com/reviewdog/reviewdog
🔎 Running pyright with reviewdog 🐶 ...
  + npm exec --yes -- pyright@latest --outputjson --lib
  No configuration file found.
  pyproject.toml file found at /home/runner/work/onnxruntime/onnxruntime.
  Loading pyproject.toml file at /home/runner/work/onnxruntime/onnxruntime/pyproject.toml
  Assuming Python version 3.10
  Assuming Python platform Linux
  No include entries specified; assuming /home/runner/work/onnxruntime/onnxruntime
  stubPath /home/runner/work/onnxruntime/onnxruntime/typings is not a valid directory.
  Searching for source files
  Found 628 source files
  An internal error occurred while type checking file "/home/runner/work/onnxruntime/onnxruntime/tools/android_custom_build/build_custom_android_package.py": TypeError: Cannot read properties of undefined (reading 'paramType')
      at map (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7982:91)
      at Array.map (<anonymous>)
      at filterOverloadMatchesForAnyArgs (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7982:44)
      at validateOverloadsWithExpandedTypes (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7879:40)
      at validateOverloadedFunctionArguments (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8138:32)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8904:48)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:3699:39)
      at doForEachSubtype (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeUtils.ts:673:9)
      at expandSubtype (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:3692:13)
      at mapSubtypesExpandTypeVars (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:3723:13)
      at validateCallArguments (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8768:28)
      at getTypeOfCall (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7374:36)
      at getTypeOfExpression (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:1022:30)
      at evaluateTypesForExpressionInContext (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:18807:21)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:866:13)
      at evaluateTypeForSubnode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:19042:9)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:865:16)
      at s.getTypeResult (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:40:20)
      at O.visitReturn (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:900:48)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:526:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at forEach (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:924:22)
      at Array.forEach (<anonymous>)
      at O.walkMultiple (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:922:15)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:917:18)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O._walkStatementsAndReportUnreachable (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:2450:18)
      at O.visitSuite (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:312:14)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:544:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O.visitFunction (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:638:18)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:442:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O._walkStatementsAndReportUnreachable (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:2450:18)
      at O.check (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:282:14)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1353:29)
      at s.timeOperation (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:44:28)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1350:45)
      at t.LogTracker.log (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/logTracker.ts:36:20)
      at t.SourceFile.check (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1348:33)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:[11](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:12)[33](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:35):40)
      at t.LogTracker.log (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/logTracker.ts:36:20)
      at L._checkTypes (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:1103:33)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:577:30)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:646:20)
      at s.runWithCancellationToken [as timeOperation] (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:44:28)
      at L._runEvaluatorWithCancellationToken (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:2467:41)
      at L.analyze (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:541:21)
      at analyzeProgram (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/analysis.ts:52:33)
      at t.BackgroundAnalysisProgram.startAnalysis (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/backgroundAnalysisProgram.ts:151:16)
      at Timeout._onTimeout (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/service.ts:1771:67)
      at listOnTimeout (node:internal/timers:559:17)
      at processTimers (node:internal/timers:502:7)
  Error performing analysis: TypeError: Cannot read properties of undefined (reading 'paramType')
      at map (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7982:91)
      at Array.map (<anonymous>)
      at filterOverloadMatchesForAnyArgs (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7982:44)
      at validateOverloadsWithExpandedTypes (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7879:40)
      at validateOverloadedFunctionArguments (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8138:32)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8904:48)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:3699:39)
      at doForEachSubtype (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeUtils.ts:673:9)
      at expandSubtype (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:3692:13)
      at mapSubtypesExpandTypeVars (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:[37](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:40)23:13)
      at validateCallArguments (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:8768:28)
      at getTypeOfCall (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:7374:36)
      at getTypeOfExpression (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:1022:30)
      at evaluateTypesForExpressionInContext (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:18807:21)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:866:13)
      at evaluateTypeForSubnode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:19042:9)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:865:16)
      at s.getTypeResult (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:40:20)
      at O.visitReturn (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:900:48)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:526:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at forEach (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:924:22)
      at Array.forEach (<anonymous>)
      at O.walkMultiple (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:922:15)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:917:18)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O._walkStatementsAndReportUnreachable (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:2450:18)
      at O.visitSuite (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:312:14)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:544:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O.visitFunction (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:6[38](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:41):18)
      at O.visit (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:442:29)
      at O.visitNode (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:933:21)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/parseTreeWalker.ts:915:37)
      at O.walk (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:303:19)
      at O._walkStatementsAndReportUnreachable (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:2450:18)
      at O.check (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/checker.ts:282:14)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1353:29)
      at s.timeOperation (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:44:28)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1350:45)
      at t.LogTracker.log (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/logTracker.ts:36:20)
      at t.SourceFile.check (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/sourceFile.ts:1348:33)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:1133:[40](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:43))
      at t.LogTracker.log (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/logTracker.ts:36:20)
      at L._checkTypes (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:1103:33)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:577:30)
      at callback (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/typeEvaluator.ts:646:20)
      at s.runWithCancellationToken [as timeOperation] (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/common/timing.ts:44:28)
      at L._runEvaluatorWithCancellationToken (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:2467:[41](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:44))
      at L.analyze (/home/runner/.npm/_npx/fbb[43](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:46)b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/program.ts:541:21)
      at analyzeProgram (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/analysis.ts:52:33)
      at t.BackgroundAnalysisProgram.startAnalysis (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/backgroundAnalysisProgram.ts:151:16)
      at Timeout._onTimeout (/home/runner/.npm/_npx/fbb43b1786f81b3f/node_modules/pyright/dist/pyright-internal/src/analyzer/service.ts:1771:67)
      at listOnTimeout (node:internal/timers:559:17)
      at processTimers (node:internal/timers:502:7)
  + true
  + python3 /home/runner/work/_actions/jordemort/action-pyright/v1/pyright_to_rdjson/pyright_to_rdjson.py
  Traceback (most recent call last):
    File "/home/runner/work/_actions/jordemort/action-pyright/v1/pyright_to_rdjson/pyright_to_rdjson.py", line 53, in <module>
      print(pyright_to_rdjson(sys.stdin))
    File "/home/runner/work/_actions/jordemort/action-pyright/v1/pyright_to_rdjson/pyright_to_rdjson.py", line 8, in pyright_to_rdjson
      pyright: Dict = json.load(jsonin)
    File "/usr/lib/python3.10/json/__init__.py", line 293, in load
      return loads(fp.read(),
    File "/usr/lib/python3.10/json/__init__.py", line 3[46](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:49), in loads
      return _default_decoder.decode(s)
    File "/usr/lib/python3.10/json/decoder.py", line 337, in decode
      obj, end = self.raw_decode(s, idx=_w(s, 0).end())
    File "/usr/lib/python3.10/json/decoder.py", line 3[55](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928568#step:9:58), in raw_decode
      raise JSONDecodeError("Expecting value", s, err.value) from None
  json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)
  + cleanup
  + '[' -n /tmp/tmp.o6rGAdR1LC ']'
  + '[' -d /tmp/tmp.o6rGAdR1LC ']'
  + rm -rf /tmp/tmp.o6rGAdR1LC
  Error: Process completed with exit code 1.
```

##### Failure 2: "Python format" failure related to "psf/black@stable": 

Many files are reported not well formatted, an example:
```
--- /home/runner/work/onnxruntime/onnxruntime/onnxruntime/python/onnxruntime_inference_collection.py	2023-02-01 03:25:08.361480 +0000
+++ /home/runner/work/onnxruntime/onnxruntime/onnxruntime/python/onnxruntime_inference_collection.py	2023-02-01 03:25:23.6[28](https://github.com/microsoft/onnxruntime/actions/runs/4060639890/jobs/6989928837#step:4:30)466 +0000
@@ -103,11 +103,10 @@
     """
     This is the main class used to run a model.
     """
 
     def __init__(self):
-
         # self._sess is managed by the derived class and relies on bindings from C.InferenceSession
         self._sess = None
         self._enable_fallback = True
 
would reformat /home/runner/work/onnxruntime/onnxruntime/onnxruntime/python/onnxruntime_inference_collection.py
     def get_session_options(self):
```

#### Root causes

Failure 1. `pyright` publish new release 1.1.292 about 4 hourse ago..
https://www.npmjs.com/package/pyright?activeTab=versions. If we revert
the version back to previous release 1.1.291, then this test pass.

Failure 2. `black` ublish its release few hours ago.
https://pypi.org/project/black/#history

![image](https://user-images.githubusercontent.com/10530022/215981502-fff0fb2f-5575-4537-aa2a-21c5a974eeb7.png)


#### Fixes

Failure 1. Fixed the `pyright` to use previous release 1.1.291.

Failure 2. This PR firstly attempted to update all impacted files based
on new version of black package offline. But we hit a throttling issue
when calling format services :
```
{"severity":"ERROR","time":"2023-02-01T08:00:08.090158864Z","logging.googleapis.com/sourceLocation":{"file":"/home/runner/work/reviewdog/reviewdog/doghouse/server/github_checker.go","line":"45","function":"github.com/reviewdog/reviewdog/doghouse/server.(*checkerGitHubClient).UpdateCheckRun"},"message":"UpdateCheckRun failed: {\"message\":\"Invalid request.\\n\\nOnly 65535 characters are allowed; 89431 were supplied.\",\"documentation_url\":\"https://docs.github.com/rest/reference/checks#update-a-check-run\"}"}
```

So an alternative fix is done here, e.g. fix the version of the black
package to previous release 22.12.0.

**Would like to get some feedback from @justinchuby, feel free to make
some change based on this to unblock other PRs.**

### 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. -->
2023-02-01 06:39:51 -08:00
Abhishek Jindal
6fa4555a06
Including support for Deepspeed 0.8.0 (#14506)
### Description
Including Support for Deepspeed 0.8.0.



### Motivation and Context
Deepspeed 0.8.0 has a bug fix and mlfow integration.
2023-02-01 06:19:41 -08:00
Baiju Meswani
d06ad9462b
[Bug Fix] Include python training apis when enable_training is enabled (#14485) 2023-01-31 17:17:26 -08:00
Erick Muñoz
d1533c27eb
[oneDNN] Improved thread handling (#13618)
* Added the OrtDnnlProviderOptions structure to expose configuration
options to the user

* The number of threads can be defined by the user with the -i flag on
the perftest

* Number of threads can also be configured via the OMP_NUM_THREADS
environment variable

* The number of threads defined in the OrtDnnlProviderOptions is
prioritized over the environment variable

### Description
Avoids thread oversubscription caused by OpenMP allocating the maximum
number of threads possible for oneDNN EP. Added support for the
OrtDnnlProviderOptions, this will allow for more EP customization
capabilities, and allows for user defined number of threads.



### Motivation and Context
- Improves performances and allows for user to fine tune the number of
threads
2023-01-31 14:37:13 -08:00
RandySheriffH
596b505f6a
Add spin args for perf test (#14507)
Add two args for spinning control for onnxruntime_perf_test:

1. Stop spinning entirely for threads in intra-op thread pool.
2. Stop spinning only between ort runs.

Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2023-01-31 14:29:53 -08:00
Yi Zhang
80f807c03d
upgrade protobuf to 3.20.2 and onnx to 1.13 (#14279)
### Description
upgrade protobuf to 3.20.2, same as onnx 1.13.0

### Motivation and Context
Per component governance requirement and Fixes #14060

unused-parameter error occurs in 2 conditions.
1. compile protolbuf

`onnxruntime_src/cmake/external/protobuf/src/google/protobuf/repeated_ptr_field.h:752:66:
error: unused parameter ‘prototype’ [-Werror=unused-parameter]`
2. include onnx_pb.h
```
2023-01-28T10:20:15.0410853Z FAILED: CMakeFiles/onnxruntime_pybind11_state.dir/onnxruntime_src/onnxruntime/python/onnxruntime_pybind_iobinding.cc.o 
......
2023-01-28T10:20:15.0466024Z                  from /build/Debug/_deps/onnx-src/onnx/onnx_pb.h:51,
2023-01-28T10:20:15.0466958Z                  from /onnxruntime_src/include/onnxruntime/core/framework/to_tensor_proto_element_type.h:10,
....
2023-01-28T10:20:15.0609678Z /build/Debug/_deps/onnx-build/onnx/onnx-operators-ml.pb.h:1178:25:   required from here
2023-01-28T10:20:15.0610895Z /onnxruntime_src/cmake/external/protobuf/src/google/protobuf/repeated_ptr_field.h:752:66: error: unused parameter ‘prototype’ [-Werror=unused-parameter]
2023-01-28T10:20:15.0611707Z cc1plus: all warnings being treated as errors

```

https://dev.azure.com/onnxruntime/2a773b67-e88b-4c7f-9fc0-87d31fea8ef2/_apis/build/builds/874605/logs/22
2023-01-31 12:55:09 -08:00
pengwa
e2dd1315c7
Fix build for --enable_language_interop_ops + DISABLE_ABSEIL=ON (#14469)
### Fix build error on Windows when building with "
--enable_language_interop_ops -cmake_extra_defines
onnxruntime_DISABLE_ABSEIL=ON"

This is a subsequent fix after
https://github.com/microsoft/onnxruntime/pull/14309, which fixed build
for onnxruntime_DISABLE_ABSEIL=ON build.

Going furthur, if we enable --enable_language_interop_ops, there are
following two errors:

```
 test_symm_qgemm.cpp
  test_transpose.cpp
onnxruntime_session.lib(inference_session.obj) : error LNK2019: unresolved external symbol "void __cdecl onnxruntime::L
oadInterOp(class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > const &,cla
ss std::vector<struct Ort::CustomOpDomain,class std::allocator<struct Ort::CustomOpDomain> > &,class std::function<void
 __cdecl(char const *)> const &)" (?LoadInterOp@onnxruntime@@YAXAEBV?$basic_string@_WU?$char_traits@_W@std@@V?$allocato
r@_W@2@@std@@AEAV?$vector@UCustomOpDomain@Ort@@V?$allocator@UCustomOpDomain@Ort@@@std@@@3@AEBV?$function@$$A6AXPEBD@Z@3
@@Z) referenced in function "public: __cdecl <lambda_f3a907e0b0a0e11d80d305605215cce8>::operator()(class std::shared_pt
r<class onnxruntime::Model> &)const " (??R<lambda_f3a907e0b0a0e11d80d305605215cce8>@@QEBA@AEAV?$shared_ptr@VModel@onnxr
untime@@@std@@@Z) [C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\onnxruntime_test_trainer.vcxproj]
onnxruntime_session.lib(inference_session.obj) : error LNK2019: unresolved external symbol "void __cdecl onnxruntime::L
oadInterOp(class onnx::ModelProto const &,class std::vector<struct Ort::CustomOpDomain,class std::allocator<struct Ort:
:CustomOpDomain> > &,class std::function<void __cdecl(char const *)> const &)" (?LoadInterOp@onnxruntime@@YAXAEBVModelP
roto@onnx@@AEAV?$vector@UCustomOpDomain@Ort@@V?$allocator@UCustomOpDomain@Ort@@@std@@@std@@AEBV?$function@$$A6AXPEBD@Z@
5@@Z) referenced in function "public: __cdecl <lambda_340b7b787b9c0f81848d348e60fe6c91>::operator()(class std::shared_p
tr<class onnxruntime::Model> &)const " (??R<lambda_340b7b787b9c0f81848d348e60fe6c91>@@QEBA@AEAV?$shared_ptr@VModel@onnx
runtime@@@std@@@Z) [C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\onnxruntime_test_trainer.vcxproj]
C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\RelWithDebInfo\onnxruntime_test_trainer.exe : fatal error
LNK1120: 2 unresolved externals [C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\onnxruntime_test_trainer.
vcxproj]
  onnxruntime.vcxproj -> C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\RelWithDebInfo\onnxruntime.dll
  onnxruntime_test_utils.vcxproj -> C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\RelWithDebInfo\onnxrun
  time_test_utils.lib
CUDACOMPILE : nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may
 be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning). [C:\Users\pengwa\dev\onnxruntime
\build\Windows\RelWithDebInfo\custom_op_library.vcxproj]
  cuda_ops.cu
CUDACOMPILE : nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may
 be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning). [C:\Users\pengwa\dev\onnxruntime
\build\Windows\RelWithDebInfo\onnxruntime_test_cuda_ops_lib.vcxproj]
```



```
  kernel_type_str_resolver_utils_test.cc
  local_kernel_registry_test.cc
C:\Users\pengwa\dev\onnxruntime\onnxruntime\test\framework\allocation_planner_test.cc(1388,9): error C2220: the followin
g warning is treated as an error [C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebInfo\onnxruntime_test_all.vcxp
roj]
C:\Users\pengwa\dev\onnxruntime\onnxruntime\test\framework\allocation_planner_test.cc(1388,9): warning C4067: unexpected
 tokens following preprocessor directive - expected a newline [C:\Users\pengwa\dev\onnxruntime\build\Windows\RelWithDebI
nfo\onnxruntime_test_all.vcxproj]
```


### 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. -->
2023-01-31 12:34:45 +08:00
Ankit
a5b620e79d [Build] Fix arm64 Docker build (#14283) 2023-01-30 16:25:19 -08:00
Wei-Sheng Chin
679ae7ff33
[Java] Fix warnings (#14076)
Fix C6011, C6385, C6386 found by Visual Studio. Basically, I set the
maximum number of options for every EP to 128. To my knowledge, 128 is
big enough to support all EPs.

For support arbitrary number of EP options, we probably need #13999 and
create a "std::vector"-like struct in C language.
2023-01-30 09:22:28 -08:00
Ashwini Khade
764202d740
fix prefast warning (#14446)
### Description
Fixes a prefast warning:
https://aiinfra.visualstudio.com/ONNX%20Runtime/_workitems/edit/11113



### 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. -->
2023-01-30 09:13:39 -08:00
cloudhan
3b6d551c35
Enable ccache for HIP objects (#14465)
This enables HIP compiler to be launched with `ccache` when build with `--use_cache`
2023-01-28 22:34:24 +08:00
Vincent Wang
7aecb2150f
Fix onnxruntime-CI-nightly-ort-pipeline Failure (#14464)
PyTorch skipped version 1.14 and jumped to 2.0, while the image for the
onnxruntime-CI-nightly-ort-pipeline is still using
nightly-ubuntu2004-cu116-py38-torch1140dev. Switch to the new torch
version image to fix the failure of the pipeline.
2023-01-28 16:05:56 +08:00
Vincent Wang
91d42e9d85
Tool to Convert ONNX Model to TFEvents (#14160)
A tool to convert ONNX model to tfevents so that we can use tensorboard
to open it for visualization. This is especially useful for debugging
when the ONNX model is too large to open by Netron.

usage: onnx2tfevents.py [-h] [--logdir LOGDIR] [--model MODEL]
2023-01-28 15:09:15 +08:00
Yulong Wang
d9219685ad
always set OpSchema in CreateNodeHelper() (#14356)
### Description
as a more generic solution to #13660, always set OpSchema in
CreateNodeHelper() so that added nodes by transformers will have
OpSchema set

### 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. -->
2023-01-27 16:56:14 -08:00
dependabot[bot]
b5b70eaa8c
Bump ua-parser-js from 0.7.31 to 0.7.33 in /js/web (#14435) 2023-01-27 23:22:48 +00:00
Zhang Lei
f87dd408f6
Support long sequence in attention (#14371)
Support long sequence in attention operator for (1) raw mask of 2/3/4-D,
(2) no mask.
Set longer greedy search max length.
2023-01-27 09:39:09 -08:00
shalvamist
368d2fc11e
Added E2E test for Image Tensor API (#14406)
### Description
Added E2E test - Currently covering -
URL --> Tensor
ImageData --> Tensor
HTML Image Element --> Tensor
Tensor --> ImageData

---------

Co-authored-by: shalvamist <shalva.mist@microsoft.com>
2023-01-27 08:54:27 -08:00
Wei-Sheng Chin
4ef64f3681
Fix warning c26409 (#14079)
We should avoid using `new` and `delete` in C/C++ code whenever possible
as suggested by VC compiler.
2023-01-26 15:43:53 -08:00
Yulong Wang
de11527d76
[js] fix js/web bundle (#14434)
### Description
make sure "crypto" is not processed by webpack for browser configuration
2023-01-26 14:43:09 -08:00
Rui Ren
eacd829d23
Bump ORT version number (#14226)
### Description
Bump ort version after the creation of release candidate of 1.14

Co-authored-by: ruiren <ruiren@microsoft.com>
2023-01-26 12:33:47 -08:00
Ye Wang
d9c744ed9a
Fix a bug in t5 beamsearch with half precision (#14436)
the CreateEncoderInputs functor was passed to the ctor as nullptr when
type is MLFloat16.

### Description
<!-- Describe your changes. -->



### 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. -->
2023-01-26 11:14:22 -08:00
liqun Fu
2b1a59f01a
cpu support of LpPool(18) (#14205)
Signed-off-by: Liqun Fu <liqfu@microsoft.com>

### Description
To support LpPool (18)



### Motivation and Context
for Ort 1.14 release

Signed-off-by: Liqun Fu <liqfu@microsoft.com>
2023-01-25 23:14:56 -08:00
Sumit Agarwal
edb377f2cb
[DML EP] Upgrade DML to 1.10.1 (#14433)
### Description
Updated DirectML version to 1.10.1
(https://www.nuget.org/packages/Microsoft.AI.DirectML/1.10.1)



### 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. -->
2023-01-25 21:07:10 -08:00
Pranav Sharma
3b8dfe2e27
Don't use free to satisfy Prefast requirements (#14354)
### Description
Don't use free to satisfy Prefast requirements

### Motivation and Context
Fix ADO#9004
2023-01-25 18:50:18 -08:00
Yulong Wang
4d9ddb5193
[js] upgrade packages in js/web/test/e2e (#14334)
### Description
upgrade versions to latest to avoid security vulerables.
2023-01-25 18:03:48 -08:00
Thiago Crepaldi
32c05fcdd1
Add Col2Im CPU op (#12311)
**Description**
This PR implements N-dimensional Col2Im as a contrib CPU Op as specified
by ONNX's https://github.com/onnx/onnx/pull/3948

**Motivation and Context**
- Col2Im enables models such as:
  - [SS-DCNet](https://github.com/xhp-hust-2018-2011/SS-DCNet)
  - [DSTT](https://github.com/ruiliu-ai/DSTT)
- It also serves to document the ORT's obscure `math::Col2ImNd` utility

Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Co-authored-by: Liqun Fu <liqfu@microsoft.com>
2023-01-25 12:23:00 -08:00
Tianlei Wu
94b1791974
Upgrade CUTLASS to v2.11 and add sequence length threshold for cutlass FMHA (#14401)
### Description
Add sequence length threshold for triggering cutlass FMHA in FP32. See
performance test results in
https://github.com/microsoft/onnxruntime/pull/14343 to see how this
threshold is selected.

Upgrade cutlass to v2.11 and update deps.txt and cgmanifest for nuget
pipeline build (test build:
https://aiinfra.visualstudio.com/Lotus/_build/results?buildId=268574&view=results)
2023-01-25 09:43:48 -08:00
Edward Chen
7cc9aed314
Android package custom build script update (#14403)
Update Android package custom build script.
- Use later version of various dependencies (CMake, JDK, Android command line tools, Android NDK, Ubuntu). The CMake version was too old for the current ORT code.
- Do in-container build in a directory that is not shared with the host. Resolves some file permission issues and speeds up file access.

Add a nightly build to make sure the script works with the latest ORT.
2023-01-25 09:19:05 -08:00
Edward Chen
3bc092b1ea
Update ORT format v5 change docs to cover limited backwards compatibility in 1.14. (#14413) 2023-01-25 08:23:12 -08:00
Adrian Lizarraga
85d7e9c596
Fix unused variable for CUDA EP builds with USE_FLASH_ATTENTION off (#14404)
### Description
Fixes unused `use_memory_efficient_attention` variable in
contrib_ops/cuda/bert/attention_impl.cu.



### Motivation and Context
ORT with CUDA version < 11.6 fails to build for release configurations
due to an unused variable.

```shell
c:\...\onnxruntime\onnxruntime\contrib_ops\cuda\bert\attention_impl.cu(420): error : variable "use_memory_efficient_attention" was declared but never referenced [C:\...\onnxruntime\build\Windows\RelWithDebInfo\onnx
runtime_providers_cuda.vcxproj]
            detected during instantiation of "onnxruntime::common::Status onnxruntime::contrib::cuda::QkvToContext(const cudaDeviceProp &, cublasHandle_t &, cudaStream_t, onnxruntime::contrib::AttentionParameters &, onnxruntime::contrib::cuda::AttentionData<T> &) [wit
  h T=float]"
  (923): here
```

This happens for CUDA < 11.6. Our cmake script turns off
onnxruntime_USE_FLASH_ATTENTION for CUDA < 11.6, which leaves the
aforementioned variable unused outside of asserts (which are removed in
release builds).

The USE_FLASH_ATTENTION option was added by
https://github.com/microsoft/onnxruntime/pull/14343
2023-01-24 09:31:57 -08:00
Edward Chen
3c1ef7dee6
Fix CI build with no Abseil. (#14400)
Use '||' instead of 'or' in onnxruntime/core/optimizer/attention_fusion_helper.h.
2023-01-24 09:17:35 -08:00
Kevin Chen
81120e9e8b
Add custom tolerance option for onnx_test_runner (#13683)
Signed-off-by: Kevin Chen <kevinch@nvidia.com>

### Description
Add a `-t` option for `onnx_test_runner` to allow users to specify
custom tolerance values when running ONNX models.


### Motivation and Context
For some backends, the default tolerance of 1-e5 is too tight to pass
accuracy checks with ONNX model zoo reference values, especially if only
one or two values are mismatched. Having a custom option will allow
different backends to specify their own custom tolerance when running
these models.

Signed-off-by: Kevin Chen <kevinch@nvidia.com>
2023-01-23 16:42:36 -08:00
liqun Fu
7b6d880b28
cpu to support bitwise ops (#14197) 2023-01-23 16:42:18 -08:00
sfatimar
77b455b969
Ort openvino 4.3 cli (#14341)
### Description
Introduce cache_dir CLI for graph serialisation.
Replace existing use_compile_network and blob_dump_path cli options for
openvino with a single command line option "cache_dir" specifying the
path that needs to be passed for blob dump/load improving the developer
experience.

### Motivation and Context?
We were having two values to set cache dir which was unnecessary

Co-authored-by: Preetha <preetha.veeramalai@intel.com>
2023-01-23 14:17:52 -08:00