feat: port + refactor ExecutorchModule to C++#345
Merged
Conversation
chmjkb
force-pushed
the
@chmjkb/executorch-module-cpp
branch
from
a1e060c to
51549bb
Compare
chmjkb
force-pushed
the
@chmjkb/executorch-module-cpp
branch
from
912dfa9 to
9f3d631
Compare
JakubGonera
suggested changes
Jun 6, 2025
chmjkb
force-pushed
the
@chmjkb/executorch-module-cpp
branch
from
4eaa14c to
3b9c932
Compare
msluszniak
reviewed
Jun 9, 2025
msluszniak
reviewed
Jun 9, 2025
JakubGonera
suggested changes
Jun 11, 2025
Comment on lines
+8
to
+25
| async load( | ||
| modelSource: ResourceSource, | ||
| onDownloadProgressCallback: (_: number) => void = () => {} | ||
| ): Promise<void> { | ||
| const paths = await ResourceFetcher.fetchMultipleResources( | ||
| onDownloadProgressCallback, | ||
| modelSource | ||
| ); | ||
| this.nativeModule = global.loadExecutorchModule(paths[0] || ''); | ||
| } | ||
|
|
||
| protected async forwardET(inputTensor: TensorPtr[]): Promise<TensorPtr[]> { | ||
| return await this.nativeModule.forward(inputTensor); | ||
| } | ||
|
|
||
| async getInputShape(methodName: string, index: number): Promise<number[]> { | ||
| return this.nativeModule.getInputShape(methodName, index); | ||
| } |
Contributor
There was a problem hiding this comment.
Two thoughts on this setup:
- The
loadmethod contains logic that would only be used if used withExecutorchModule(correct me if I'm wrong), so we don't gain anything on having this code here compared to inExecutorchModule. - Adding
forwardETto the base class requires each module inheriting to be a single model, so it excludes things like OCR which will not have a singleforwardETmethod we could invoke. Maybe we should inherit fromExecutorchModulewhere possible (e.g.StyleTransfer), and have this be a true base class for all modules?
Collaborator
Author
There was a problem hiding this comment.
- That's correct, this setup was done strictly for testing purposes and I forgot to change it
- Yeah, that makes sense but I'm not sure if this can be done now. Currently the native API accepts one source arg on load. So either we refactor the native API and then make the changes you mentioned, or we can do something like this:
class EncoderDecoderModel { // doesn't inherit from base, kind of like a controller
constructor(encoderSource, decoderSource) {
// they do inherit from base
this.encoder = new SomeEncoder(encoderSource)
this.decoder = new SomeDecoder(decoderSource)
}
}
msluszniak
reviewed
Jun 11, 2025
chmjkb
force-pushed
the
@chmjkb/executorch-module-cpp
branch
from
97ed56d to
bc75e3a
Compare
JakubGonera
approved these changes
Jun 12, 2025
mkopcins
pushed a commit
that referenced
this pull request
Oct 15, 2025
## Description This PR aims to replace the existing ExecuTorch bindings for ones that leverage JSI and resemble the underlying runtime more accurately. Currently WIP. ### Type of change - [ ] Bug fix (non-breaking change which fixes an issue) - [ ] New feature (non-breaking change which adds functionality) - [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected) - [ ] Documentation update (improves or adds clarity to existing documentation) ### Tested on - [ ] iOS - [ ] Android ### Testing instructions <!-- Provide step-by-step instructions on how to test your changes. Include setup details if necessary. --> ### Screenshots <!-- Add screenshots here, if applicable --> ### Related issues <!-- Link related issues here using #issue-number --> #209 #338 - [ ] I have performed a self-review of my code - [ ] I have commented my code, particularly in hard-to-understand areas - [ ] I have updated the documentation accordingly - [ ] My changes generate no new warnings ### Additional notes <!-- Include any additional information, assumptions, or context that reviewers might need to understand this PR. -->
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
This PR aims to replace the existing ExecuTorch bindings for ones that leverage JSI and resemble the underlying runtime more accurately. Currently WIP.
Type of change
Tested on
Testing instructions
Screenshots
Related issues
#209
#338
Additional notes