When I have a cluster already running, I sometimes want to re-run the setup commands even if there's no changes, without having to shut down the cluster. For example if I am installing a package via a Github repo, I want to trigger the install again after pushing new changes.

However, calling ray up doesn

PrestoDB https://prestodb.io .. is widely used as SQL frontend for many different data-sources, including ElasticSearch, and even files in S3 .. would be very nice if there would be a Connector available for Vespa.

Currently the way that the predict function converts the incoming data and more importantly the return statement is quite obscure, and the community tends to just look at the util.py functions, however they shouldn't have the need to read the source to know what rules apply. For this it would be useful to ahve a table or a high level overview of the rules that apply in regards to what data types g

Problem description

This is a request for a new feature on CI testing.

We would like to have a larger coverage for converter CI tests.
You may refer here https://github.com/microsoft/hummingbird/tree/master/tests for traditional ML -> {PyTorch, ONNX} first.

Other Information

Roadmap (you may change this as you prefer)