fix(ml): armnn not being used #10929
Merged
Conversation
This file contains 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
fyfrey
approved these changes
Jul 10, 2024
claabs
pushed a commit
to claabs/immich-machine-learning-openvino-kernel-fix
that referenced
this pull request
Aug 11, 2024
* fix armnn not being used, move fallback handling to main, add tests * formatting
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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
We check if an ARM NN model exists and fall back to ONNX if it does not. However, we do this check before the model is downloaded at all, so it will always appear missing unless it has already been downloaded. Additionally, since we have already fallen back to ONNX when downloading, we avoid downloading the ARM NN model. This catch-22 prevents ARM NN from being used in practice without manually moving the model into the model cache.
This PR fixes this behavior and additionally moves the fallback handling outside of the model class itself to being in the main
loadfunction. This makes the model's behavior more predictable: it's unexpected for the model to choose a different format than what the caller specified.