Repository navigation
Fingerprint's YAMNet is fetched once on every machine (for 0.1.3) - #140
Merged
Merged
Conversation
…ystems/columns models.rs), from the public repo's models-1 release (Google's model, Apache-2.0, as ONNX; perception's copy, sha256 8e4e21d5) and checked against that sha256, so every machine has the same encoders. Before, only machines with perception's copy had YAMNet: a fingerprint trained on one (Shane's) refused a fresh install's columns ('trained on other encoders'), so a fingerprint kept with a project didn't judge on a teammate's machine. A fingerprint trained without YAMNet takes it as an encoder added since. raea-sys-ml models.rs gains ensure_url, a file from a URL of its own
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Only machines with perception's hand-made YAMNet copy had the sound-events encoder. A fingerprint records the encoders it was trained on and refuses a machine with a different set, so a fingerprint trained on Shane's Mac (which pins yamnet/yamnet.onnx) failed on a fresh install with 'trained on other encoders'. That breaks 0.1.3's Keep with this project for teammates.
SigLIP 2 and CLAP already match Hugging Face's files byte for byte (the pins in Shane's fingerprint are HF's LFS sha256s), so YAMNet was the only difference. A fingerprint trained without YAMNet takes it as an encoder added since (verbs.rs open), so 0.1.2 users' fingerprints keep working.
Branched from rc3 (3784608) so rc4 is rc3 plus this. Checked after rc4 builds: its raea with an empty RAEA_MODELS fetches YAMNet and judges with Shane's fingerprint.