Add initial support for model metadata in RTen models #48
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
Add a new `Metadata` table to the FlatBuffers schema and `Model::metadata` structure for accessing this data in Rust. The initial set of fields includes a description, license and metadata describing the provenance of the model: - The SHA-256 hash of the input ONNX model - The code repository (eg. GitHub) - The model repository (eg. Hugging Face) - Training run ID and URL (eg. from an experiment tracking platform such as Weights and Biases)
This initial support assumes that the metadata is stored separately from the ONNX file. In future it will also be useful to import from metadata stored within the ONNX file itself.
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Outdated suggestions cannot be applied.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.
Add initial support for storing model metadata in RTen models. Metadata is accessed via
Model::metadatain Rust and added to converted models by specifying a JSON file containing the metadata when runningconvert-onnx.py, using a new--metadataflag.The ONNX spec is quite bare-bones with respect to standard metadata fields. RTen is a bit more prescriptive about what should be included and the formats of those fields.
Summary of changes:
Metadatatable to FlatBuffers schema for.rtenmodelsMetadatatable when converting ONNX modelsModel::metadatato access model metadatarten-cli <model_path>Part of #20