The release of the pruning code

[LoggerHead22]

Hi, are you going to release the pruning code of the model? will it be released as part of the modelopt library?

I found a pull request with an example of pruning llama-3.1-8b with mode="mcore_gpt_minitron", that uses import modelopt.torch.prune as mtp, but the current version of modelopt does not have this module.

Perhaps you can tell me where to find the source code of the model pruning using the method of the Minitron article, or when to expect it (and is it worth expecting at all)?