Jack Gerrits

No, you're welcome to work on this

Hi @dnabanita7, yes this is still open. Please feel free to work on it :)

This issue is a proposal that needs to be discussed before any work should be done on it. Taking a look at issues labeled as [`good first issue`](https://github.com/VowpalWabbit/vowpal_wabbit/issues?q=is%3Aissue+is%3Aopen+label%3A%22Good+First+Issue%22) is a...

There is also this PR which seems relevant: #3194

For extreme multiclass classification Probabalistic Label Tree is a good option, used with `--plt`. You can learn more about how to use it here: https://github.com/VowpalWabbit/vowpal_wabbit/tree/master/demo/plt However, the number of classes...

It looks like `plt` and `log_multi` both use a sort of hierarchical approach. But I am not sure if they are the same as what you are thinking.

Honestly, I'm not sure. I don't know if they've been tested at that scale.

https://stackoverflow.com/a/685422

There is no pre-canned solution for "what do I do if I have data that was collected with exploration but I don't have the probabilities?". Some people try and learn...

Nugets are produced per commit. These nugets contain `vw.exe`. https://github.com/VowpalWabbit/vowpal_wabbit/actions/workflows/build_nugets.yml @lokitoth are you satisfied this is resolved?