Commit graph

651 commits

Author SHA1 Message Date
Wanming Lin
41ad83fb00
[WebNN EP] Support rest Reduction ops for TFLite backend (#21135)
- reduceLogSum, reduceLogSumExp and reduceSumSquare have been landed in
https://chromium-review.googlesource.com/c/chromium/src/+/5575815
- reduceL1 and reduceL2 have been landed in
https://chromium-review.googlesource.com/c/chromium/src/+/5606091
2024-06-25 18:30:55 -07:00
Wanming Lin
4743803944
[WebNN EP] Support more Normalization ops for TFLite backend (#21151)
Following Normalization ops have been supported in Chromium for TFLite
backend:
- batchNormalization:
https://chromium-review.googlesource.com/c/chromium/src/+/5532745
- layerNormalization:
https://chromium-review.googlesource.com/c/chromium/src/+/5573326
- instanceNormalization:
https://chromium-review.googlesource.com/c/chromium/src/+/5532750
2024-06-24 19:04:23 -07:00
Wanming Lin
3a917e49fb
[WebNN EP] Support 4 more ops for TFLite backend (#21134)
Recently WebNN TFLite backend supports gelu, expand, softsign,
reciprocal.
2024-06-24 09:52:12 -07:00
Wanming Lin
0c80cd2157
[WebNN EP] Update Prelu restriction for CPU backend (#20878) 2024-06-20 11:04:01 -07:00
Xu Xing
c3076721f3
[js/webgpu] Support conv3d naive (#20706)
### 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. -->
2024-06-19 10:13:50 -07:00
Wanming Lin
40879a2623
[WebNN EP] Enable Cast op for WebNN CPU backend (#20864)
WebNN TFLite backend supports `cast` op but doesn't support casting to
`uint64` data type.
2024-06-19 01:51:19 -07:00
Wanming Lin
35c430a95a
[WebNN EP] Enable several ops for WebNN CPU backend (#20847)
WebNN CPU implementation has been migrated from XNNPack to TFLite which
supports more ops. Turn on partial `cpu` supported ops which just need
the change from `false` to `true` firstly.
2024-06-19 01:45:31 -07:00
Yulong Wang
5e81fa8aec
[js] fix vulnerability CVE-2024-4068: upgrade braces to 3.0.3 (#21078)
### Description

Upgrade `braces` to 3.0.3

[CVE-2024-4068](https://github.com/advisories/GHSA-grv7-fg5c-xmjg)

```
# npm audit report

braces  <3.0.3
Severity: high
Uncontrolled resource consumption in braces - https://github.com/advisories/GHSA-grv7-fg5c-xmjg
fix available via `npm audit fix`
node_modules/braces

1 high severity vulnerability
```
2024-06-18 16:02:08 -07:00
Yulong Wang
631a2c16be
[js/web] skip default locateFile() when dynamic import is disabled (#21073)
### Description
skip default `locateFile()` when dynamic import is disabled. This allows
the file to work with bundlers to load WebAssembly file correctly if
`env.wasm.wasmPaths` is not set.
2024-06-18 12:21:45 -07:00
Yang Gu
1473d66a00
[js/webgpu] Prefer adapter.info to adapter.requestAdapterInfo (#21065)
WebGPU is deprecating async adapter.requestAdapterInfo, and replacing it
with sync adapter.info.
Spec change: https://github.com/gpuweb/gpuweb/pull/4662
2024-06-18 12:02:38 -07:00
Jian Chen
4e18b0b7ce
Upgrade braces from 3.0.2 to 3.0.3 to fix the vulnerability (#21022) 2024-06-12 18:02:52 -07:00
Yulong Wang
dd805ff77d
[js/web] ESM: use the bundled target as default export (#20991)
### Description
ESM: use the bundled target as default export

In this change, the default import of the following entries:
```
import from 'onnxruntime-web';
import from 'onnxruntime-web/all';
import from 'onnxruntime-web/webgpu';
```
will use the "bundled" version, which has no dynamic import.

This change should only apply to ESM on web.
2024-06-11 11:14:55 -07:00
Wanming Lin
043ef5c95f
[WebNN EP] Support latest WebNN softmax op (#20827)
Latest WebNN softmax supports N-D input and axis parameter.
2024-06-11 08:27:14 -07:00
Edward Chen
981893c318
Remove deprecated "mobile" packages (#20941)
# Description

This PR removes the building of the ORT "mobile" packages and much of the associated infrastructure which is no longer needed.

Not removed yet - tools/ci_build/github/android/mobile_package.required_operators.config and the helper scripts that depend on it.

# Motivation and Context

The mobile packages were deprecated in 1.18. Users should use the full packages (Android - onnxruntime-android, iOS - onnxruntime-c/onnxruntime-objc) instead or do a custom build.
2024-06-07 16:20:32 -05:00
Wanming Lin
52874f628a
[WebNN EP] Remove some constraints for CPU backend (#20900)
Following constraints have been supported by WebNN TFLite backend:
- Concat: supports up to 4 inputs
- Matmul: supports broadcasting
- Resize: supports nearest mode
- Split: supports up to 4 outputs
2024-06-06 08:22:41 -07:00
Wanming Lin
da1f8f9274
[WebNN EP] TFLite backend only supports limit ranges for Clip (#20863) 2024-06-06 08:22:18 -07:00
Guenther Schmuelling
c749bd997a
webgpu quickgelu (#20939) 2024-06-06 08:21:33 -07:00
Changming Sun
3dd6fcc089
Upgrade min ios version to 13.0 (#20773)
To align with Office and other MS products.
Office's support policy is:
"Office for iPad and iPhone is supported on the two most recent versions
of iOS and iPadOS. When a new version of iOS or iPadOS is released, the
Office Operating System requirement becomes the two most recent
versions: the new version of iOS or iPadOS and the previous version."
(from https://products.office.com/office-system-requirements)

The latest iOS version is 17. So they support both 17 and 16. Here I set
our min iOS version to 13 so that it will be a superset of what Office
supports.

This change would allow us using C++17's std::filesystem feature in the
core framework. The modifications were generated by running
```bash
 find . -type f -exec sed -i "s/apple_deploy_target[ =]12.0/apple_deploy_target=13.0/g"  {} \;
```

Cannot use 15.0 because otherwise iOS packaging would fail with:

```
/Users/runner/work/1/b/apple_framework/intermediates/iphoneos_arm64/Release/_deps/coremltools-src/mlmodel/src/MILBlob/Util/Span.hpp:288:9: error: cannot use 'throw' with exceptions disabled
        MILVerifyIsTrue(index < Size(), std::range_error, "index out of bounds");
```

The Google OSS libraries we use only officially support iOS 15+.
2024-06-04 10:15:20 -07:00
Wanming Lin
9c6481fa2d
[WebNN EP] Enable ArgMax and ArgMin for CPU backend (#20865)
WebNN TFLite backend supports ArgMax and ArgMin, but only supports
'select_last_index' value is 0.
2024-06-03 14:12:11 -07:00
Wanming Lin
c128132dd8
[WebNN EP] TFLite backend only supports Elu with default alpha (#20862) 2024-06-03 14:10:22 -07:00
Yulong Wang
ab9f153746
[js/web] allow build target for non dynamic import (#20898)
### Description
<!-- Describe your changes. -->

This PR allows to build ORT web to `ort{.all|.webgpu}.bundle.min.mjs`,
which does not have any dynamic import. This makes it possible to use
ort web via static import in service worker.

Fixes #20876
2024-06-03 12:33:37 -07:00
Yulong Wang
35697d2421
[js/webnn] update API of session options for WebNN (#20816)
### Description

This PR is an API-only change to address the requirements being
discussed in #20729.

There are multiple ways that users may create an ORT session by
specifying the session options differently.

All the code snippet below will use the variable `webnnOptions` as this:
```js
const myWebnnSession = await ort.InferenceSession.create('./model.onnx', {
   executionProviders: [
     webnnOptions
   ]
});
```

### The old way (backward-compatibility)

```js
// all-default, name only
const webnnOptions_0 = 'webnn';

// all-default, properties omitted
const webnnOptions_1 = { name: 'webnn' };

// partial
const webnnOptions_2 = {
  name: 'webnn',
  deviceType: 'cpu'
};

// full
const webnnOptions_3 = {
  name: 'webnn',
  deviceType: 'gpu',
  numThreads: 1,
  powerPreference: 'high-performance'
};
```

### The new way (specify with MLContext)

```js
// options to create MLcontext
const options = {
  deviceType: 'gpu',
  powerPreference: 'high-performance'
};

const myMlContext = await navigator.ml.createContext(options);

// options for session options
const webnnOptions = {
  name: 'webnn',
  context: myMlContext,
  ...options
};
```

This should throw (because no deviceType is specified):
```js
const myMlContext = await navigator.ml.createContext({ ... });
const webnnOptions = {
  name: 'webnn',
  context: myMlContext
};
```

### Interop with WebGPU
```js
// get WebGPU device
const adaptor = await navigator.gpu.requestAdapter({ ... });
const device = await adaptor.requestDevice({ ... });

// set WebGPU adaptor and device
ort.env.webgpu.adaptor = adaptor;
ort.env.webgpu.device = device;

const myMlContext = await navigator.ml.createContext(device);
const webnnOptions = {
  name: 'webnn',
  context: myMlContext,
  gpuDevice: device
};
```

This should throw (because cannot specify both gpu device and MLContext
option at the same time):
```js
const webnnOptions = {
  name: 'webnn',
  context: myMlContext,
  gpuDevice: device,
  deviceType: 'gpu'
};
```
2024-05-31 03:25:14 -07:00
Xu Xing
25ac65375c
[js/webgpu] Fix mha name (#20860)
### 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. -->
2024-05-30 00:01:06 -07:00
Peishen Yan
cfe68e489e
[WebNN EP] Support Trilu op (#20730)
Adds support for Trilu via WebNN Triangular op
2024-05-24 10:46:54 -07:00
Guenther Schmuelling
33a68d221f
add missing file for pr20791 (#20811)
this file should have been in pr20791 to allow fp16 in the tile
implementation
2024-05-24 09:59:13 -07:00
Satya Kumar Jandhyala
bab5037eab
Eliminate explicit Concat operations in Attention (#20556)
### Description
Remove explicitly concatinating pastKey with Key and pastValue with
Value.



### 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. -->
2024-05-24 09:07:57 -07:00
Wanming Lin
2c39d0c502
[WebNN EP] Disable ConvTranspose for WebNN CPU (#20762)
WebNN CPU backend implementation has been migrated from XNNPack to
TFLite, currently TFLite has not supported WebNN's convTranspose2d yet,
just disable it for now.
2024-05-22 20:59:37 -07:00
Yulong Wang
e412bc1919
[doc] update file size table for ORT Web (#20755)
### 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. -->
2024-05-22 11:04:57 -07:00
Xu Xing
f1fef19b6e
[js/webgpu] Support shared memory for transpose 2d (#19267)
For 1024x1024, without shared memoey, 18.7ms. With shared memory 13.2ms.
2024-05-22 08:15:44 -07:00
Yulong Wang
068bb3d5ee
[js/webgpu] add missing space in build script (#20752) 2024-05-21 16:24:34 -07:00
Wanming Lin
87d49e3dda
[WebNN EP] Add WebNN operators doc to README.md (#20734) 2024-05-20 14:57:40 -07:00
Wanming Lin
0399d1b12d
[WebNN EP] Update chromium flag (#20732)
WebNN is currently enabled behind "Enables WebNN API" flag.
2024-05-20 14:57:30 -07:00
Yulong Wang
036fcd93d4
[js/web] optimize module export and deployment (#20165)
### Description

This PR make numbers of optimizations to onnxruntime-web's module export
and deployment.

See each section below for more details.

#### Preview

>
[onnxruntime-web@1.19.0-esmtest.20240513-a16cd2bd21](https://www.npmjs.com/package/onnxruntime-web/v/1.19.0-esmtest.20240513-a16cd2bd21)

> ~~onnxruntime-web@1.19.0-esmtest.20240430-c7edbcc63d~~

> ~~onnxruntime-web@1.18.0-esmtest.20240428-624c681c83~~

> ~~onnxruntime-web@1.18.0-esmtest.20240411-1abb64e894~~

<details>
<summary><h4>Breaking changes</h4></summary>

There is no code change required, but there are a few differences
regarding **code import**, **flags**, **bundler config** and
**deployment steps**.

#### Importing:

Import table is changed. See following for details.

<details>
<summary><h5>Current import table:</h5></summary>

| Target Name | Path for "import" or "require" | WebGL | JSEP | wasm |
Proxy | Training |
  |------|-----|-----|-----|-----|-----|-----|
  | `ort` (default) | `onnxruntime-web` | ✔️ |  | ✔️ | ✔️ |  |
  | `ort.all` | `onnxruntime-web/experimental` | ✔️ | ✔️ | ✔️ | ✔️ |  |
  | `ort.node` | `onnxruntime-web` |  |  | ✔️ |  |  |
| `ort.training` | `onnxruntime-web/training` |  |  | ✔️ |
✔️<sup>\[1]</sup> | ✔️ |
  | `ort.wasm` | `onnxruntime-web/wasm` |  |  | ✔️ | ✔️ |  |
  | `ort.wasm-core` | `onnxruntime-web/wasm-core` |  |  | ✔️ |  |  |
| `ort.webgl` | `onnxruntime-web/webgl` | ✔️ |  |  | ✔️<sup>\[2]</sup>
|  |
  | `ort.webgpu` | `onnxruntime-web/webgpu` |  | ✔️ | ✔️ | ✔️ |  |

* [1] didn't test. may not actually work.
* [2] not working. this is a mistake in build config.

</details>

<details>
<summary><h5>Proposed update:</h5></summary>

| Target Name | Path for "import" or "require" | WebGL | JSEP | wasm |
Proxy | Training |
  |------|-----|-----|-----|-----|-----|-----|
  | `ort` (default) | `onnxruntime-web` | ✔️ |  | ✔️ | ✔️ |  |
| `ort.all` |
~~`onnxruntime-web/experimental`~~<br/>`onnxruntime-web/all` | ✔️ | ✔️ |
✔️ | ✔️ |  |
  | `ort.node` | `onnxruntime-web` |  |  | ✔️ |  |  |
  | `ort.training` | `onnxruntime-web/training` |  |  | ✔️ | ✔️ | ✔️ |
  | `ort.wasm` | `onnxruntime-web/wasm` |  |  | ✔️ | ✔️ |  |
| ~~`ort.wasm-core`~~ | ~~`onnxruntime-web/wasm-core`~~ | ~~~~ | ~~~~
| ~~✔️~~ | ~~~~ | ~~~~ |
  | `ort.webgl` | `onnxruntime-web/webgl` | ✔️ |  |  | ~~✔️~~  |  |
  | `ort.webgpu` | `onnxruntime-web/webgpu` |  | ✔️ | ✔️ | ✔️ |  |

</details>

#### Flags:

The following flags are deprecated:
- `env.wasm.simd` (boolean): will be ignored. SIMD is always enabled in
build.

The following flags changed their type:
- `env.wasm.wasmPaths`: When using this flag as a string ( for the URL
prefix ), nothing is changed. When using this flag as an object ( for
per-file path override ), the type changed:
  ```diff
  -  export interface Old_WasmFilePaths{
  -    'ort-wasm.wasm'?: string;
  -    'ort-wasm-threaded.wasm'?: string;
  -    'ort-wasm-simd.wasm'?: string;
  -    'ort-training-wasm-simd.wasm'?: string;
  -    'ort-wasm-simd-threaded.wasm'?: string;
  -  };
  +  export interface New_WasmFilePaths {
  +    /**
  +     * Specify the override path for the main .wasm file.
  +     *
  +     * This path should be an absolute path.
  +     *
  +     * If not modified, the filename of the .wasm file is:
  +     * - `ort-wasm-simd-threaded.wasm` for default build
+ * - `ort-wasm-simd-threaded.jsep.wasm` for JSEP build (with WebGPU and
WebNN)
  +     * - `ort-training-wasm-simd-threaded.wasm` for training build
  +     */
  +    wasm?: URL|string;
  +    /**
  +     * Specify the override path for the main .mjs file.
  +     *
  +     * This path should be an absolute path.
  +     *
  +     * If not modified, the filename of the .mjs file is:
  +     * - `ort-wasm-simd-threaded.mjs` for default build
+ * - `ort-wasm-simd-threaded.jsep.mjs` for JSEP build (with WebGPU and
WebNN)
  +     * - `ort-training-wasm-simd-threaded.mjs` for training build
  +     */
  +    mjs?: URL|string;
  +  }
  ```

#### Bundler compatibility:

Config changes are need for bundlers. See usage example in
/js/web/test/e2e/ for Webpack, parcel and rollup.

#### Deployment:

- if consuming from a CDN, there is no breaking change.
- if consuming from a local server, need to copy all `ort-*.wasm` and
`ort-*.mjs` files (totally 6 files) in the dist folder. (previously only
need to copy `ort-*.wasm` files.)

</details>
<details>
<summary><h4>Problems</h4></summary>

There are a few problems with the current module export and deployment:

- Script URL cannot be correctly inferred when imported as ESM.
- Workers are forcefully encoded using Blob URL, which makes
onnxruntime-web not working in CSP environment and Node.js, when using
proxy or multi-threading feature.
- Generated JS code (by Emscripten) is encoded using
`function.toString()`, which is unstable and error-prone.
- When running with a different Emscripten build, always need the build
step. Making it difficult to swap artifacts in deveopment/debug.
</details>
<details>
<summary><h4>Goals</h4></summary>

- Full ESM support
- Support variances of ways to import. Including:
- import from HTML's `<script>` tag (IIFE format, exporting to global
variable `ort`)
    ```html
<script
src="https://example.com/cdn-path-to-onnxruntime-web/dist/ort.min.js"></script>
    ```
  - import from source code inside `<script type="module">` tag (ESM)
    ```html
    <script type="module">
import * as ort from
"https://example.com/cdn-path-to-onnxruntime-web/dist/ort.min.mjs";

      // using 'ort'
    </script>
    ```
- import in a CommonJS project (CJS format, resolve from package.json
"exports" field)
    ```js
    // myProject/main.js
    const ort = require('onnxruntime-web');
    ```
- import in an ESM project (ESM format, resolve from package.json
"exports" field)
    ```js
    // myProject/main.js (or main.mjs)
    import * as ort from 'onnxruntime-web';
    ```
- Support popular bundlers when importing onnxruntime-web into a CJS/ESM
project.
  - webpack (esm requires extra post-process step)
  - rollup
  - parcel (esm requires extra post-process step)
  - More bundlers **TBD**
- Multi-threading support for Node.js

NOTE: keeping single JavaScript file (the all-in-one bundle) is no
longer a goal. This is because technically there is a conflict with the
other requirements.
</details>

<details>
<summary><h4>Important Design Decisions</h4></summary>

- Drop support of single JavaScript output.
- The current onnxruntime-web distribution uses a single JavaScript file
to include all code. While there are a few benefits, it also creates
problems as mentioned above. Since ESM is being used more and more
widely, and browsers are making more restricted security checks and
requirement, the old Blob based solution is going to be replaced.
- To achieve the requirement, specifically, the CSP environment support,
we have to offer a non Blob based solution. Therefore, we have to
distribute multiple files and drop the single file solution.

- Do not run parser/postprocess on Emscripten generated JavaScript.
- Emscripten is evolving quickly so we should only depends on what's in
its documentation instead of a certain implementation details. (for
example, currently we patch on its code to deal with a special variable
`_scriptDir`)
  - Keep the generated files as-is also helps to:
    - reduce the size of ort.min.js
- make it easier to replace build artifacts when in development/debug

- Drop support for non-SIMD and non-MultiThread. This helps to reduce
the number of artifacts in distribution.
  - (fixed-sized) SIMD is supported in any mainstream JS environment.
- Multi-thread as WebAssembly feature is supported in any mainstream JS
environment. In some environment the feature is guarded with cross
origin policy, but it can still work if not trying to create any worker.

- Use ESM output for Emscripten generated JavaScript.
- There are 2 ways to dynamically import classic (umd) modules and
neither of them are recommended:
- dynamically creating a <script> tag. This changes the HTML structure
and have quite a lot of compatibility issue
- use `fetch()` and `eval()`. However `eval` is strongly suggested to be
avoid because there is a great perf hit.
- importing ESM is super easy - just use the `import()` call.
Considering ESM is widely supported in modern browsers and Node.js this
is the better option.

- Add Blob based solution as a fallback for cross-origin workers.
- There are still wide use case of importing onnxruntime-web from CDN.
In this usage, make it able create worker by using `fetch()`+`Blob` to
create a same-origin Blob URL.

</details>

<details>
<summary><h4>Distribution File Manifest</h4></summary>

The distribution folder contains the following files:

- WebAssembly artifacts. These files are the result of compiling the
ONNX Runtime C++ code to WebAssembly by Emscripten.

  | File Name | Build Flags |
  |------|-----|
| ort-wasm-simd-threaded.mjs <br/> ort-wasm-simd-threaded.wasm |
`--enable_wasm_simd` <br/> `--enable_wasm_threads` |
| ort-training-wasm-simd-threaded.mjs <br/>
ort-training-wasm-simd-threaded.wasm | `--enable_training_apis` <br/>
`--enable_wasm_simd` <br/> `--enable_wasm_threads` |
| ort-wasm-simd-threaded.jsep.mjs <br/> ort-wasm-simd-threaded.jsep.wasm
| `--enable_wasm_simd` <br/> `--enable_wasm_threads` <br/> `--use_jsep`
<br/> `--use_webnn` |

- onnxruntime-web JavaScript artifacts. These files are generated by
ESBuild as the entry point for onnxruntime-web.

  There are multiple build targets for different use cases:
  | Target Name | Path for "import" or "require" | Description |
  |------|-----|-----|
  | `ort` | `onnxruntime-web` | The default target. |
  | `ort.all` | `onnxruntime-web/all` | The target including webgl. |
  | `ort.node` | `onnxruntime-web` | The default target for Node.js. |
| `ort.training` | `onnxruntime-web/training` | The target including
training APIs |
| `ort.wasm` | `onnxruntime-web/wasm` | The target including only
WebAssembly (CPU) EP |
| `ort.webgl` | `onnxruntime-web/webgl` | The target including only
WebGL EP |


  For each target, there are multiple files generated:
  | File Name | Description |
  |------|-----|
| [target].js | The entry point for the target. IIFE and CommonJS
format. |
  | [target].mjs | The entry point for the target. ESM format. |
| [target].min.js <br/> [target].min.js.map | The entry point for the
target. Minimized with sourcemap. IIFE and CommonJS format. |
| [target].min.mjs <br/> [target].min.mjs.map | The entry point for the
target. Minimized with sourcemap. ESM format. |
| [target].proxy.mjs | (if appliable) The proxy ESM module for the
target. |
| [target].proxy.min.mjs <br/> [target].proxy.min.mjs.map | (if
appliable) The proxy ESM module for the target. Minimized with
sourcemap. |

</details>

<details>
<summary><h4>Dynamic Import Explained</h4></summary>

- Local Served | No Proxy:
  ```
  [Bundle or ort.min.js]
    |
    + import()--> [ort-wasm-simd-threaded.mjs]
                    |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
                    |
+ new Worker()--> [ort-wasm-simd-threaded.mjs (worker)]
                                        |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
  ```
- Local Served | Proxy:
  ```
  [Bundle or ort.min.js]
    |
    + import()--> [ort.proxy.min.mjs]
                    |
                    + new Worker()--> [ort.proxy.min.mjs (worker)]
                                        |
+ import()--> [ort-wasm-simd-threaded.mjs]
                                                        |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
                                                        |
+ new Worker()--> [ort-wasm-simd-threaded.mjs (worker)]
|
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
  ```
- Cross Origin | No Proxy:
  ```
  [Bundle or ort.min.js]
    |
    + fetch('ort-wasm-simd-threaded.mjs')
        |
        + URL.createObjectURL(res.blob())
        |
        + import()--> [blob:... (ort-wasm-simd-threaded)]
                        |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
                        |
+ new Worker()--> [blob:... (ort-wasm-simd-threaded) (worker)]
                                            |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
  ```

- Cross Origin | Proxy
  ```
  [Bundle or ort.min.js]
    |
    + fetch('ort.proxy.min.mjs')
        |
        + URL.createObjectURL(res.blob())
        |
        + import()--> [blob:... (ort.proxy)]
                        |
+ new Worker()--> [blob:... (ort.proxy) (worker)]
                                            |
+ fetch('ort-wasm-simd-threaded.mjs')
                                                |
+ URL.createObjectURL(res.blob())
                                                |
+ import()--> [blob:... (ort-wasm-simd-threaded)]
                                                                |
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
                                                                |
+ new Worker()--> [blob:... (ort-wasm-simd-threaded) (worker)]
|
+ WebAssembly.instantiateStreaming()--> [ort-wasm-simd-threaded.wasm]
  ```
</details>
2024-05-20 09:51:16 -07:00
Hans
2a17958b34
[js/rn] Fix some bugs (#20242)
### Description
<!-- Describe your changes. -->
- Fix `logSeverityLevel`
- Correct get RCTCxxBridge, old method for some cases will got wrong
bridge


### 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: Yulong Wang <7679871+fs-eire@users.noreply.github.com>
2024-05-15 10:32:08 -07:00
Xu Xing
8c59cd4fce
[js/webgpu] Support GroupQueryAttention (#20237)
TODOs:
1. Handle H * params.kvNumHeads greater than work group size limit.
2. Support BNSH kv cache.
2024-05-13 09:43:37 -07:00
Jon Campbell
768c79317c
Enable QNN HTP support for Node (#20576)
### Description
Add support for using Onnx Runtime with Node

### Motivation and Context
Onnx Runtime supports the QNN HTP, but does not support it for Node.js.
This adds baseline support for the Onnx Runtime to be used with Node.

Note it does not update the node packages that are distributed
officially. This simply patches the onnxruntime.dll to allow 'qnn' to be
used as an execution provider.

Testing was done using the existing onnxruntime-node package. The
`onnxruntime.dll` and `onnxruntime_binding.node` were swapped into
`node_modules\onnxruntime-node\bin\napi-v3\win32\arm64` with the newly
built version, then the various QNN dlls and .so files were placed next
to the onnxruntime.dll. Testing was performed on a variety of models and
applications, but the easiest test is to modify the [node quickstart
example](https://github.com/microsoft/onnxruntime-inference-examples/tree/main/js/quick-start_onnxruntime-node).
2024-05-09 13:11:07 -07:00
Guenther Schmuelling
55a6986d38
optimize skiplayernorm (#20551)
SkipSimplifiedLayerNormalization used in phi3 comes down from 222usec to
14usec
2024-05-08 08:40:03 -07:00
Yi-Hong Lyu
b2481e3602
Bump up version in main from 1.18.0 to 1.19.0 (#20489)
Bump up version in main from 1.18.0 to 1.19.0 since the release branch
has been cut.

---------

Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
2024-04-29 20:21:41 -07:00
Yulong Wang
b1085b51ca
[js/web] update README (#20492)
### Description

Update README.md in /js/web/

- update compatibility table
- update links to onnxruntime.ai
2024-04-29 17:56:23 -07:00
Satya Kumar Jandhyala
99b0e19f11
[JS/WebGPU] MatMulNBits remove unnecessary condition (#20396)
Distribute writing-to-output work over all threads in MatMulNBits.
### 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. -->
2024-04-29 14:27:21 -07:00
Satya Kumar Jandhyala
736cbb3925
[JS/WebGU] Support fp16 in Attention by performing the computation in fp32. (#20486)
### Description
Perform computation in fp32 and convert finally to fp16.



### 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. -->
2024-04-27 08:30:26 -07:00
Satya Kumar Jandhyala
21b3cbc3af
[WIP][JS/WebGPU] Inputs Key and Value could be 4-dims. (#20470)
### Description
The Key and Value inputs could be 4-dims


### 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. -->
2024-04-25 13:33:46 -07:00
Yulong Wang
a5182a2ef3
[js/web] update test condition for '--force-localhost' (#20450)
### Description

Fixes the NPM packaging pipeline failure.
2024-04-24 12:14:03 -07:00
Satya Kumar Jandhyala
ae78cdb5d7
[JS/WebGPU] MultiheadAttention bugfix (#20447)
### Description
Fixed pastkey, key and pastvalue, value concatenation condition and
fixed index error. Added new test cases.



### 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. -->
2024-04-24 08:43:14 -07:00
Guenther Schmuelling
33d5ea39b3
[js/webgpu] fixes for fp16 attention (#20440) 2024-04-24 08:01:28 -07:00
Yulong Wang
8f53957bcf
[js/web] add "browser" field to support parcel v2 (#20422)
### Description

As described in latest discussion in #19915, parcel v2 without using the
[new resolver](https://parceljs.org/blog/v2-9-0/#new-resolver) will not
work correctly with onnxruntime-web. There are still users who uses
parcel with default resolver, so add this deprecated field "browser"
back for backward compatibility. This PR also corrects the "main" field,
which is for old resolver for Node.js.
2024-04-23 13:10:11 -07:00
Yulong Wang
13bda11583
[Node.js binding] Fix install script (#20416)
### Description
Fix a few bugs of the install script of onnxruntime-node package.

This change is integrated from branch `rel-1.17.3` (#20397)
2024-04-23 13:01:16 -07:00
Satya Kumar Jandhyala
d42ac7f0c6
[JS/WebGPU] Multihead attention improvements (#20286)
### Description
Enabled more usecases



### 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. -->
2024-04-23 12:39:49 -07:00
Guenther Schmuelling
b8e6684313
more conservitive gpu-buffer cache algo (#20312)
tuned based on 80 models to keep performance impact minimal
2024-04-23 09:07:04 -07:00
Yulong Wang
4385602386
[js/web] fix test runner with optional input/output (#20399)
### Description
fix test runner with optional input/output.

This change fixes the OP test runner (.jsonc format test) with optional
input(s) and/or output(s).

this fix reveals a problem of dealing with optional outputs:

> Take SkipSimplifiedLayerNorm as example: 
>
> if in the ONNX model, the node's outputs are: [ 'output_0', '' ]
instead of [ 'output_0' ], the current implementation will fail. The
difference is, in the first case, context.outputCount == 2, and then the
typescript implementation will try to create a tensor for output[1]. It
will eventually call to C++ function (OpKernelContext::Output), and the
output.DataRaw() will be nullptr. WebGPU backend will fail because it
cannot deal with a TensorView with data == 0.
>

This problem may need to be fixed or workaround in separated PR. This PR
does not fix this problem. Failed test cases are modified to work -
please note this PR does not break those test cases as they never work.
2024-04-22 12:53:10 -07:00