Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature Request] Include and expose protobuf serialization / deserialization in the JS interface (ideally in onnxruntime-common) #21297

Open
wbeardall opened this issue Jul 9, 2024 · 0 comments
Labels
feature request request for unsupported feature or enhancement platform:web issues related to ONNX Runtime web; typically submitted using template

Comments

@wbeardall
Copy link

Describe the feature request

For certain use-cases, I find myself wanting to pass around serialized representations of data efficiently, both for caching and efficient transfer between machines. As such, exposing protobuf-based serialization and deserialization in the JS interfaces, in a manner similar to the helpers from the Python library, would be very beneficial. As-is, there does not seem to be a straightforward way to convert ONNX-serialized objects to their JS interface version, and vice versa, directly in the interface.

Describe scenario use case

This feature would be very useful for handling short-term and intermediate-term caching of tensor objects during runtime, as well as passing serialized tensors across different machines or environments. This would be particularly useful for cases where passing tensors efficiently between Python and JS environments is required.

@wbeardall wbeardall added the feature request request for unsupported feature or enhancement label Jul 9, 2024
@snnn snnn added the platform:web issues related to ONNX Runtime web; typically submitted using template label Jul 9, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature request request for unsupported feature or enhancement platform:web issues related to ONNX Runtime web; typically submitted using template
2 participants