Commit graph

7820 commits

Author SHA1 Message Date
Changming Sun
070769d61d
Use onnxruntime_fetchcontent_makeavailable cmake function for TRT (#13918)
### Description
Use onnxruntime_fetchcontent_makeavailable cmake function for TRT. See
the comment for the reason.


### Motivation and Context
To support a newer TRT version. Previously they have a "BUILD_EXE" build
option to allow us to exclude such things from build. But in
https://github.com/onnx/onnx-tensorrt/pull/879 they deleted the build
option. It wouldn't be a problem if we continue to use git submodules as
before, because cmake's add_subdirectories function has an
"EXCLUDE_FROM_ALL" keyword. However, cmake's FetchContent module
doesn't. That's why I needed to create our own version of the macro.
2022-12-12 11:27:46 -08:00
RandySheriffH
75584c5fa8
Enabling thread pool to be numa-aware (#13778)
The PR enables ort thread pool to be numa-aware, so that threads could
be evenly created and distributed among numa nodes.
In addition, to facilitate performance tuning, the PR opens a new API
allowing customers to attach threads to certain logical processors.
Please check the API
[definition](https://github.com/microsoft/onnxruntime/pull/13778/files#diff-5845a5c76fb64abdc8f0cffe21b37f8da1712674eb3abc4cd87190891be1bd48)
for details.

Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2022-12-12 10:33:55 -08:00
Jian Chen
b8d941f065
Cjian/pad ops bug (#13930) 2022-12-12 10:23:49 -08:00
Sumit Agarwal
fe827c3891
[DML EP] Disable DML Graph Fusion for lower graph optimization level OR setOptimizedFilePath true (#13913)
### Description
DML EP won't fuse the ONNX Graph if ORT Graph optimization level is <= 1
or `SessionOption::SetOptimizedFilePath` is passed.

This is the successor of
https://github.com/microsoft/onnxruntime/pull/11346.

### Motivation and Context
- **Why is this change required? What problem does it solve?**  
Requested by few a users (issues below) and also helps in debugging.
- **If it fixes an open issue, please link to the issue here:**
  - https://github.com/microsoft/onnxruntime/issues/13535
  - https://github.com/microsoft/onnxruntime/issues/8440
2022-12-12 10:15:51 -08:00
Edward Chen
8cfbc4fe91
Add support for other data types to Split CPU kernel. (#13900)
Split copies data - we can add support for all data types without too much binary size impact by using data type size-based implementations. The DispatchStridedCopy() function used here does this.
2022-12-12 09:29:15 -08:00
Yi Zhang
2cb12caf93
Output cache stats (#13937)
### Description
Output cache stats
2022-12-12 15:22:13 +08:00
Changming Sun
89812a623e
Add two daily build jobs to validate some extra build configs (#13921)
### Description
Add two daily build jobs to validate some extra build configs


### Motivation and Context

To catch issues like: #13893
2022-12-10 09:15:14 -08:00
Hariharan Seshadri
51aaf2e021
Allow using separate GPT2 decoder subgraphs for the initial run and the subsequent runs in BeamSearch/GreedySearch (#13914) 2022-12-10 08:02:35 -08:00
JiCheng
22fa62152a
Pass SessionOptions to XnnpackProviderFactoryCreator. (#13318)
### Description
To pass session_options to Xnnpack EP via
`XnnpackProviderFactoryCreator` for Initializing xnnpack's threadpool.

If you want to use different threadpool size or even disable xnnpack's
threadpool, just setting intra_threadpool to 1 by xnnpack EP's
provider_options.


### Motivation and Context

Co-authored-by: Guangyun Han <guangyunhan@microsoft.com>
Co-authored-by: Jicheng Wen <jicwen@microsoft.com>
2022-12-10 14:23:46 +08:00
Edward Chen
87eef1fe21
Use updated ONNX license in ThirdPartyNotices.txt. (#13919)
Use updated ONNX license in ThirdPartyNotices.txt. It got changed to the Apache license.
Copied LICENSE file content from onnx submodule at cmake/external/onnx.
2022-12-09 17:46:37 -08:00
Ashwini Khade
a7bc927b4b
fix typos in training apis (#13908)
### Description
This PR fixes some typos in the training apis.

We need to add more tests and make sure they are all run on the CIs to
capture such issues. These changes are out of scope of this PR.



### 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. -->

Co-authored-by: Ashwini Khade <askhade@microsoft.com@orttrainingdev8.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>
2022-12-09 16:01:11 -08:00
shalvamist
1921d84636
Updated ORT-Web build instructions (#13282)
### Description
Replaced the previous build steps with the latest documentation in
onnxruntime.ai

### Motivation and Context
Removing duplicates in the documentation sources
2022-12-09 15:58:39 -08:00
Nat Kershaw (MSFT)
21dd341e52
Add Google Analytics to python apidocs (#13901) 2022-12-09 15:44:12 -08:00
Adrian Lizarraga
db9c677b63
[EP Perf Dashboard] Add TensorRT 8.5.1.1 dockerfile (#13843)
### Description
- Adds a dockerfile for Ubuntu with TensorRT 8.5.1.1.
- Adds option to run EP Perf pipeline with TensorRT 8.5

### Motivation and Context
Necessary to benchmark models with TensorRT 8.5
2022-12-09 14:33:52 -08:00
Abhishek Udupa
83c59d2594
Session-aware and thread-safe CUDA profiler (#13706)
### Description
The existing CUDA profiler is neither session-aware, nor thread-safe.
This PR ensures both.

### Motivation and Context
[PR 13549](https://github.com/microsoft/onnxruntime/pull/13549) brought
thread-safety and session-awareness to the ROCm profiler. This PR brings
the same goodness to the CUDA profiler as well.

Sample outputs of a profiling run from the StableDiffusion model (this
model was chosen because it requires orchestration of multiple sessions,
and verifies that the profilers are now indeed session-aware) on both
CUDA and ROCm EPs are attached, along with a script that checks that the
trace files generated by the profile are well-formed.

Update 11/29: Updated the profile outputs. The older profile outputs
exhibited an issue where some timestamps were wildly out of range,
leading to problems visualizing the traces. The bug has been fixed and
the profile outputs have been updated, along with an update to the check
script to ensure that timestamps are monotonically increasing.


[sd_profile_outputs_cuda.tar.gz](https://github.com/microsoft/onnxruntime/files/10118088/sd_profile_outputs_cuda.tar.gz)

[sd_profile_outputs_rocm.tar.gz](https://github.com/microsoft/onnxruntime/files/10118089/sd_profile_outputs_rocm.tar.gz)

[check_profile_output_well_formedness.zip](https://github.com/microsoft/onnxruntime/files/10118090/check_profile_output_well_formedness.zip)

Co-authored-by: Abhishek Udupa <abhishek.udupa@microsoft.com>
2022-12-09 13:22:12 -08:00
dependabot[bot]
18d5cd6ee5
Bump Newtonsoft.Json from 13.0.1 to 13.0.2 in /csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests.Mobile/EndToEndTests.Mobile.Automation (#13884)
[//]: # (dependabot-start)
⚠️  **Dependabot is rebasing this PR** ⚠️ 

Rebasing might not happen immediately, so don't worry if this takes some
time.

Note: if you make any changes to this PR yourself, they will take
precedence over the rebase.

---

[//]: # (dependabot-end)

Bumps [Newtonsoft.Json](https://github.com/JamesNK/Newtonsoft.Json) from
13.0.1 to 13.0.2.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/JamesNK/Newtonsoft.Json/releases">Newtonsoft.Json's
releases</a>.</em></p>
<blockquote>
<h2>13.0.2</h2>
<ul>
<li>New feature - Add support for DateOnly and TimeOnly</li>
<li>New feature - Add UnixDateTimeConverter.AllowPreEpoch property</li>
<li>New feature - Add copy constructor to JsonSerializerSettings</li>
<li>New feature - Add JsonCloneSettings with property to disable copying
annotations</li>
<li>Change - Add nullable annotation to JToken.ToObject(Type,
JsonSerializer)</li>
<li>Change - Reduced allocations by reusing boxed values</li>
<li>Fix - Fixed MaxDepth when used with ToObject inside of a
JsonConverter</li>
<li>Fix - Fixed deserializing mismatched JToken types in properties</li>
<li>Fix - Fixed merging enumerable content and validate content</li>
<li>Fix - Fixed using $type with arrays of more than two dimensions</li>
<li>Fix - Fixed rare race condition in name table when deserializing on
device with ARM processors</li>
<li>Fix - Fixed deserializing via constructor with ignored base type
properties</li>
<li>Fix - Fixed MaxDepth not being used with ISerializable
deserialization</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="4fba53a324"><code>4fba53a</code></a>
Remove prerelease for 13.0.2</li>
<li><a
href="b15df4b50d"><code>b15df4b</code></a>
Add missing headers</li>
<li><a
href="789bfd3bbc"><code>789bfd3</code></a>
Update to 13.0.2-beta3</li>
<li><a
href="b13717a1c1"><code>b13717a</code></a>
Add JsonCloneSettings to disable copy annotations (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2757">#2757</a>)</li>
<li><a
href="d0a328e8a4"><code>d0a328e</code></a>
Fix MaxDepth not being used with ISerializable deserialization (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2736">#2736</a>)</li>
<li><a
href="aae9284e20"><code>aae9284</code></a>
Update SDK</li>
<li><a
href="bd989708b1"><code>bd98970</code></a>
Update to 13.0.2-beta2</li>
<li><a
href="4dc9af66e0"><code>4dc9af6</code></a>
Add roll forward to global.json (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2726">#2726</a>)</li>
<li><a
href="b8f4ef0f98"><code>b8f4ef0</code></a>
Fixing misspelling (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2698">#2698</a>)</li>
<li><a
href="cb9eed9666"><code>cb9eed9</code></a>
Fix deserializing via constructor with ignored base type properties (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2711">#2711</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/JamesNK/Newtonsoft.Json/compare/13.0.1...13.0.2">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=Newtonsoft.Json&package-manager=nuget&previous-version=13.0.1&new-version=13.0.2)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
- `@dependabot use these labels` will set the current labels as the
default for future PRs for this repo and language
- `@dependabot use these reviewers` will set the current reviewers as
the default for future PRs for this repo and language
- `@dependabot use these assignees` will set the current assignees as
the default for future PRs for this repo and language
- `@dependabot use this milestone` will set the current milestone as the
default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/microsoft/onnxruntime/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-12-09 13:07:05 -08:00
dependabot[bot]
6f25f6e1f0
Bump Newtonsoft.Json from 13.0.1 to 13.0.2 in /csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests (#13885)
[//]: # (dependabot-start)
⚠️  **Dependabot is rebasing this PR** ⚠️ 

Rebasing might not happen immediately, so don't worry if this takes some
time.

Note: if you make any changes to this PR yourself, they will take
precedence over the rebase.

---

[//]: # (dependabot-end)

Bumps [Newtonsoft.Json](https://github.com/JamesNK/Newtonsoft.Json) from
13.0.1 to 13.0.2.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/JamesNK/Newtonsoft.Json/releases">Newtonsoft.Json's
releases</a>.</em></p>
<blockquote>
<h2>13.0.2</h2>
<ul>
<li>New feature - Add support for DateOnly and TimeOnly</li>
<li>New feature - Add UnixDateTimeConverter.AllowPreEpoch property</li>
<li>New feature - Add copy constructor to JsonSerializerSettings</li>
<li>New feature - Add JsonCloneSettings with property to disable copying
annotations</li>
<li>Change - Add nullable annotation to JToken.ToObject(Type,
JsonSerializer)</li>
<li>Change - Reduced allocations by reusing boxed values</li>
<li>Fix - Fixed MaxDepth when used with ToObject inside of a
JsonConverter</li>
<li>Fix - Fixed deserializing mismatched JToken types in properties</li>
<li>Fix - Fixed merging enumerable content and validate content</li>
<li>Fix - Fixed using $type with arrays of more than two dimensions</li>
<li>Fix - Fixed rare race condition in name table when deserializing on
device with ARM processors</li>
<li>Fix - Fixed deserializing via constructor with ignored base type
properties</li>
<li>Fix - Fixed MaxDepth not being used with ISerializable
deserialization</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="4fba53a324"><code>4fba53a</code></a>
Remove prerelease for 13.0.2</li>
<li><a
href="b15df4b50d"><code>b15df4b</code></a>
Add missing headers</li>
<li><a
href="789bfd3bbc"><code>789bfd3</code></a>
Update to 13.0.2-beta3</li>
<li><a
href="b13717a1c1"><code>b13717a</code></a>
Add JsonCloneSettings to disable copy annotations (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2757">#2757</a>)</li>
<li><a
href="d0a328e8a4"><code>d0a328e</code></a>
Fix MaxDepth not being used with ISerializable deserialization (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2736">#2736</a>)</li>
<li><a
href="aae9284e20"><code>aae9284</code></a>
Update SDK</li>
<li><a
href="bd989708b1"><code>bd98970</code></a>
Update to 13.0.2-beta2</li>
<li><a
href="4dc9af66e0"><code>4dc9af6</code></a>
Add roll forward to global.json (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2726">#2726</a>)</li>
<li><a
href="b8f4ef0f98"><code>b8f4ef0</code></a>
Fixing misspelling (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2698">#2698</a>)</li>
<li><a
href="cb9eed9666"><code>cb9eed9</code></a>
Fix deserializing via constructor with ignored base type properties (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2711">#2711</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/JamesNK/Newtonsoft.Json/compare/13.0.1...13.0.2">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=Newtonsoft.Json&package-manager=nuget&previous-version=13.0.1&new-version=13.0.2)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
- `@dependabot use these labels` will set the current labels as the
default for future PRs for this repo and language
- `@dependabot use these reviewers` will set the current reviewers as
the default for future PRs for this repo and language
- `@dependabot use these assignees` will set the current assignees as
the default for future PRs for this repo and language
- `@dependabot use this milestone` will set the current milestone as the
default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/microsoft/onnxruntime/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-12-09 13:06:26 -08:00
dependabot[bot]
fba49952d2
Bump Newtonsoft.Json from 13.0.1 to 13.0.2 in /csharp/test/Microsoft.ML.OnnxRuntime.Tests.Common (#13886)
Bumps [Newtonsoft.Json](https://github.com/JamesNK/Newtonsoft.Json) from
13.0.1 to 13.0.2.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/JamesNK/Newtonsoft.Json/releases">Newtonsoft.Json's
releases</a>.</em></p>
<blockquote>
<h2>13.0.2</h2>
<ul>
<li>New feature - Add support for DateOnly and TimeOnly</li>
<li>New feature - Add UnixDateTimeConverter.AllowPreEpoch property</li>
<li>New feature - Add copy constructor to JsonSerializerSettings</li>
<li>New feature - Add JsonCloneSettings with property to disable copying
annotations</li>
<li>Change - Add nullable annotation to JToken.ToObject(Type,
JsonSerializer)</li>
<li>Change - Reduced allocations by reusing boxed values</li>
<li>Fix - Fixed MaxDepth when used with ToObject inside of a
JsonConverter</li>
<li>Fix - Fixed deserializing mismatched JToken types in properties</li>
<li>Fix - Fixed merging enumerable content and validate content</li>
<li>Fix - Fixed using $type with arrays of more than two dimensions</li>
<li>Fix - Fixed rare race condition in name table when deserializing on
device with ARM processors</li>
<li>Fix - Fixed deserializing via constructor with ignored base type
properties</li>
<li>Fix - Fixed MaxDepth not being used with ISerializable
deserialization</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="4fba53a324"><code>4fba53a</code></a>
Remove prerelease for 13.0.2</li>
<li><a
href="b15df4b50d"><code>b15df4b</code></a>
Add missing headers</li>
<li><a
href="789bfd3bbc"><code>789bfd3</code></a>
Update to 13.0.2-beta3</li>
<li><a
href="b13717a1c1"><code>b13717a</code></a>
Add JsonCloneSettings to disable copy annotations (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2757">#2757</a>)</li>
<li><a
href="d0a328e8a4"><code>d0a328e</code></a>
Fix MaxDepth not being used with ISerializable deserialization (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2736">#2736</a>)</li>
<li><a
href="aae9284e20"><code>aae9284</code></a>
Update SDK</li>
<li><a
href="bd989708b1"><code>bd98970</code></a>
Update to 13.0.2-beta2</li>
<li><a
href="4dc9af66e0"><code>4dc9af6</code></a>
Add roll forward to global.json (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2726">#2726</a>)</li>
<li><a
href="b8f4ef0f98"><code>b8f4ef0</code></a>
Fixing misspelling (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2698">#2698</a>)</li>
<li><a
href="cb9eed9666"><code>cb9eed9</code></a>
Fix deserializing via constructor with ignored base type properties (<a
href="https://github-redirect.dependabot.com/JamesNK/Newtonsoft.Json/issues/2711">#2711</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/JamesNK/Newtonsoft.Json/compare/13.0.1...13.0.2">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=Newtonsoft.Json&package-manager=nuget&previous-version=13.0.1&new-version=13.0.2)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
- `@dependabot use these labels` will set the current labels as the
default for future PRs for this repo and language
- `@dependabot use these reviewers` will set the current reviewers as
the default for future PRs for this repo and language
- `@dependabot use these assignees` will set the current assignees as
the default for future PRs for this repo and language
- `@dependabot use this milestone` will set the current milestone as the
default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/microsoft/onnxruntime/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-12-09 13:04:18 -08:00
Changming Sun
d5b45226be
Improve the handling of /external:I (#13904)
### Description

Improve the handling of "/external:I". The
"onnxruntime_external_lib_include_dir" variable may be:

1. A simple file path
2. A cmake generator expression like "$<INSTALL_INTERFACE:include>",
"$<TARGET_PROPERTY:onnx_proto,INTERFACE_INCLUDE_DIRECTORIES>",
"$<BUILD_INTERFACE:xxxx>". It seems that we can't simply put them in to
the "target_compile_options" line. So this PR tries to parse the
expression and extract the part we need out.

### Motivation and Context
Resolve the Github issue: https://github.com/microsoft/onnxruntime/issues/13893
2022-12-09 11:44:32 -08:00
Edward Chen
d8e22f6e50
Update VerifyOutputs() to use SpanEq() instead of gsl::span comparison operators which may be disabled. (#13911) 2022-12-09 11:31:09 -08:00
Rachel Guo
dead5c6b3a
Revert "[js/rn] support load model from buffer on Android (#12676)" (#13903)
### Description
<!-- Describe your changes. -->

As title.

This pr is missing an un-updated index.android.gradle, which causing an
unstable e2e unit test run for React Native CI.

Revert the changes for now.

### 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. -->

To unblock Ort React Native CI pipeline failure.
2022-12-09 11:05:54 -08:00
shalvamist
d22be84add
Pin packaging to version 21.3 to address training pipeline failures 2022-12-09 09:05:55 -08:00
Changming Sun
05dc1165a5
Add protobuf version constraint (#13870)
To fix a build error:


/home/xxxxxxxxxxxxx/onnxruntime/build/Linux/Debug/tensorboard/compat/proto/cost_graph.pb.cc:17:8:
error:
‘PROTOBUF_INTERNAL_EXPORT_tensorboard_2fcompat_2fproto_2ftensor_5fshape_2eproto’
does not name a type
17 | extern
PROTOBUF_INTERNAL_EXPORT_tensorboard_2fcompat_2fproto_2ftensor_5fshape_2eproto
::PROTOBUF_NAMESPACE_ID::internal::SCCInfo<1>
scc_info_TensorShapeProto_tensorboard_2fcompat_2fproto_2ftensor_5fshape_2eproto;
2022-12-08 16:14:16 -08:00
Adam Louly
fb4707f76d
add cuda support to python bindings (#13700)
### Description
Add cuda support to the on device training python bindings.



### Motivation and Context
Now users can set the execution provider (cpu or cuda) when using python
bindings for on device training apis.
2022-12-08 16:03:53 -08:00
Abhishek Udupa
7d684d1255
Include algorithm selection exposed by ROCBLAS extensions API in GEMM autotuning (#13831)
### Description
Extend GEMM autotuning by including algorithms exposed by a ROCBLAS
extension API.

### Motivation and Context
Based on our request, the ROCm team has implemented extension APIs in
ROCBLAS that provides a list of application GEMM
algorithms/implementations for a given input size, along with an API
that actually performs the GEMM using the specified
implementation/algorithm. We have observed that the ROCBLAS
algorithm/implementation selection logic does not always pick the
optimal. This PR uses the extension APIs to integrate the exposed
ROCBLAS algorithms/implementations into the autotuning framework.

The feature is disabled by default (the ROCBlas extension APIs are
slated to be released with ROCm 5.5, and are not yet generally
available). To enable: build with `--cmake-extra-defines
USE_ROCBLAS_EXTENSION_API=1 CMAKE_HIP_FLAGS=-DUSE_ROCBLAS_EXTENSION_API`
and then enable tuning in the provider options.

Co-authored-by: Abhishek Udupa <abhishek.udupa@microsoft.com>
2022-12-08 14:21:17 -08:00
Yulong Wang
dbf47284d1
[wasm] disable closure compiler in debug build (#13865)
### Description
disable closure compiler in debug build. after this change, emscripten
will only run closure compiler in release build.
2022-12-08 13:18:19 -08:00
Changming Sun
81c2defd3b
Remove unused git submodules (#13830) 2022-12-07 21:59:16 -08:00
PeixuanZuo
c1cc1d5859
[ROCm] Update FastGelu and add kernel expolrer test for FastGeluStaticSelection (#13758)
### Description
<!-- Describe your changes. -->

1. Update FastGelu conditions for supported parameters, avoid redundant
configurations participating in tuning。
2. Add kernel explorer test for FastGeluStaticSelection

### 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. -->

Co-authored-by: peixuanzuo <peixuanzuo@linmif39a000004.zvflicr54joexhdgnhvmxrxygg.phxx.internal.cloudapp.net>
2022-12-08 12:37:10 +08:00
PeixuanZuo
7694b695a9
[ROCm] Simplify ROCm manylinux dockerfile (#13873)
### Description
<!-- Describe your changes. -->

1. Remove ROCm5.3 pipeline because it has rocblas bug, we don't need it.
2. We removed the dependency on centos docker image provided by
AMD(https://hub.docker.com/r/rocm/dev-centos-7) and build ROCm centos
base image by ourselves. The reference
dockerfile(https://github.com/RadeonOpenCompute/ROCm-docker/blob/master/dev/Dockerfile-centos-7)
is very redundant for our need. We simplified the ROCm manylinux
dockerfile.
3. Different versions of rocm use the same dockerfile
`Dockerfile.manylinux2014_rocm`.

### 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. -->

Co-authored-by: peixuanzuo <peixuanzuo@linmif39a000004.zvflicr54joexhdgnhvmxrxygg.phxx.internal.cloudapp.net>
2022-12-08 09:18:27 +08:00
Edward Chen
a64ddb36d0
Always build with XNNPACK EP in iOS CI build. (#13850)
Always build with XNNPACK EP in iOS CI build.
Combine builds for CPU, CoreML, and XNNPACK EPs due to limited build agent resources.
2022-12-07 16:08:34 -08:00
Sumit Agarwal
5b16593192
[DML EP] Attention Kernel bug fix (#13879)
### Description
- Use same data type as input for mask_index tensor which is used as DML
GEMM API's C parameter.
- Remove gsl header include as it is already gets included transitively.



### Motivation and Context
- Why is this change required? What problem does it solve?
Bug found in internal conformance testing.
- If it fixes an open issue, please link to the issue here.
N/A
2022-12-07 15:24:27 -08:00
Yulong Wang
4c79977f52
[wasm] fix session option setting of mem_pattern (#13858)
### Description
 fix session option setting of memory pattern.
2022-12-07 13:15:44 -08:00
dependabot[bot]
ffdcde7cc7
Bump minimatch from 3.0.4 to 3.0.5 in /js/web (#13722)
Bumps [minimatch](https://github.com/isaacs/minimatch) from 3.0.4 to
3.0.5.
<details>
<summary>Commits</summary>
<ul>
<li><a
href="707e1b231d"><code>707e1b2</code></a>
3.0.5</li>
<li><a
href="a8763f4388"><code>a8763f4</code></a>
Improve redos protection, add many tests</li>
<li><a
href="bafa295617"><code>bafa295</code></a>
Use master branch for travis badge</li>
<li><a
href="013d64dc24"><code>013d64d</code></a>
update travis</li>
<li>See full diff in <a
href="https://github.com/isaacs/minimatch/compare/v3.0.4...v3.0.5">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=minimatch&package-manager=npm_and_yarn&previous-version=3.0.4&new-version=3.0.5)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
- `@dependabot use these labels` will set the current labels as the
default for future PRs for this repo and language
- `@dependabot use these reviewers` will set the current reviewers as
the default for future PRs for this repo and language
- `@dependabot use these assignees` will set the current assignees as
the default for future PRs for this repo and language
- `@dependabot use this milestone` will set the current milestone as the
default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/microsoft/onnxruntime/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-12-07 13:14:59 -08:00
Adam Louly
f453d2845e
adding get and set lr for optimizer (#13661)
### Description
Exposing get and set Learning rate for optimizer


### Motivation and Context
you can now set learning rate for optimizer.
2022-12-07 11:59:11 -08:00
Ashwini Khade
983877c712
Decouple strided tensor support from ENABLE_TRAINING (#13829)
### Description
Decouple strided tensor support from ENABLE_TRAINING

### Motivation and Context
This is step 1 for creating a dedicated build for on device training.
Intention is

1. We can set ENABLE_STRIDED_TENSORS in cmake when either
ENABLE_TRAINING or ENABLE_TRAINING_ON_DEVICE is selected, this way we
dont have to use if defined(ENABLE_TRAINING) ||
defined(ENABLE_TRAINING_ON_DEVICE ) everywhere in the code.

2. This also paves the way to easily enable strided tensor support for
inference in future (if required).
2022-12-07 09:22:21 -08:00
Yi Zhang
f6c493793d
Revert "skip TestCUDAProviderOptions in End2EndTest (#13737)" (#13874)
This reverts commit 87d5703b14.

### Motivation and Context
There was a bug in Linux CUDA installation.
The OS image is updated.
The TestCUDAProviderOptions could be reenabled.
2022-12-07 23:33:59 +08:00
Yi Zhang
ae2a9373ab
reenable quant model tests (#13871)
### Description

### Motivation and Context
Test data in the image has been fixed.
2022-12-07 23:33:22 +08:00
Patrice Vignola
96d8d2c278
[DML EP] Add SkipLayerNormalization (#13849)
### Description

Add SkipLayerNormalization for the DML EP
2022-12-07 01:49:14 -08:00
Hariharan Seshadri
004a1538d3
Extend vocab padding for logits MatMul for fp16 GPT2 GreedySearch (#13842) 2022-12-06 19:39:20 -08:00
cloudhan
f79d38181b
Fix hipify to avoid nccl_service.h: No such file or directory (#13852)
Fix various flaky build error due to onnxruntime_session missing dependencies on hipify generated files.
2022-12-07 09:10:37 +08:00
Changming Sun
d12521d7b2
Upgrade pybind11 (#13853)
Upgrade pybind11 to include the fix for #9735
2022-12-06 15:39:23 -08:00
Yi Zhang
78d18fbf34
Use CacheTask to Accelerate MacOS build (#13859)
### Description
Use CCache and ADO CacheTask to Accelerate MacOS build.
ref:
https://learn.microsoft.com/en-us/azure/devops/pipelines/release/caching?view=azure-devops

### Motivation and Context
The MacOS CI duration could be reduced from more than **70minutes** to
**10 minutes**

https://dev.azure.com/onnxruntime/onnxruntime/_build/results?buildId=824912&view=results
2022-12-07 07:14:40 +08:00
Yi Zhang
d2188fbff9
skip resnet50-int8 model test in training (#13856)
### 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. -->
2022-12-06 22:47:24 +08:00
Ashwini Khade
65201e47bf
Enable nuget packages for on device training (#13637)
### Description
This PR enables building nuget packages locally for on device training
using --build_nuget arg.
This PR also enables the C# bindings by default in the managed package.
If a user triggers any training apis when the native binary is not built
for training, an exception with message "Training is disabled in the
current build. Please build ONNXRuntime from source with the build flags
enable_training and enable_training_on_device. " is thrown.

Build command for creating nuget packes for on device training:
build.bat --enable_training --enable_training_on_device --build_nuget 

2 Nuget packages are built
1. Microsoft.ML.OnnxRuntime.Managed
2. Microsoft.ML.OnnxRuntime.Training OR
Microsoft.ML.OnnxRuntime.Training.Gpu



### 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. -->
2022-12-05 14:54:09 -08:00
JiCheng
d5574e6999
LayerNorm test fix (#13840)
### Description
<!-- Describe your changes. -->

Testcases of LayerNorm with fp16/bf16 are failed in Andriod and IOS
since the two platforms don't support the combinations of datatypes as
well.


https://dev.azure.com/onnxruntime/onnxruntime/_build?definitionId=134&_a=summary

https://dev.azure.com/onnxruntime/onnxruntime/_build?definitionId=53&_a=summary

### 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. -->
2022-12-05 22:49:22 +08:00
Hariharan Seshadri
5f4e0c95ec
Misc minor bug fixes in transformer kernels (#13780) 2022-12-04 21:30:57 -08:00
mindest
f34ebbc8ff
fix a wrong assert condition in benchmark_helper (#13821)
### Description
fix a wrong assert condition in benchmark_helper.py (introduced in
#13455)
2022-12-03 18:50:47 +08:00
Pranav Sharma
335b62bde6
Fix invocation of GetInputMemoryType. (#13828)
### Description
GetInputMemoryType was introduced in ver 13 in [this
PR](https://github.com/microsoft/onnxruntime/pull/10879). The ver check
introduced in this PR allows custom ops compiled using older versions to
work with newer versions (> 12) of the ORT binary.

### Motivation and Context
Fixes binary compatibility.
2022-12-02 18:42:14 -08:00
Patrice Vignola
b53bbe7370
[DML EP] Add an implementation for NonZero (#13768)
### Description
Add the NonZero op for DML



### Motivation and Context
NonZero is used in a few transformer models, so having a DML
implementation will stop large tensors from being transferred to the CPU
and back to the GPU
2022-12-02 18:39:21 -08:00
Gaz Iqbal
b9702587df
[oneDNN] Implemented Concat Op (#13646)
### Description
This PR implements the **Concat Operator** for the **OneDNN Execution
Provider**.

### Motivation and Context
- As part of evaluating ORT performance on ARM based targets such as
Graviton3, we discovered that the OneDNN EP had some gaps on operator
coverage.
- The Concat Operator is fairly common and used in models such as
Yolov5, MobileNet, DistillBert and GPT2
- For Yolov5 specifically, this improves average inference time over 100
runs on Graviton3 from 180.2ms to 115.5ms when using OneDNN + ARM
Compute Library.

Co-authored-by: Gaz Iqbal <giqbal@octoml.ai>
2022-12-02 13:30:37 -08:00