mlem VS Ray

Mlem for Lemmy has a similar look and feel to Apollo. Still missing a lot of features though. https://github.com/iterative/mlem

Here are a few things to consider: 1. You're using both DVC and MLflow to store the model artifact, why? 2. How I envision MLflow, DVC and git to work together is like this. DVC to manage the training dataset, git to manage the code, and MLflow will do the rest. About the part about "versioning" the model, MLflow has a model registry feature to "tag" a well-performing experiment. 3. Or just to do everything in DVC. DVC also has a way to do experiment tracking. Then if you need a model registry there's MLEM by the same company.

No, it's a completely separate open source tool, not directly related to DVC - https://github.com/iterative/mlem

This functionality can be used from open source tool mlem.ai and our released UI - https://studio.iterative.ai/

MLEM is a new MLOps tool to bridge the gap between ML engineers and DevOps teams by using the git-based approach that developers are already familiar with. Using MLEM, developers can store and track their ML models throughout their lifecycle: GitHub - iterative/mlem: 🐶 Version and deploy your ML models following GitOps principles

22. Ray | Github | tutorial

Training times for GSM8k are mentioned here: https://github.com/ray-project/ray/tree/master/doc/source/te...

Here's a reference from RLlib that points to docs and an example, and here's one from one of my projects that includes all my own implementations

RLlib

⦿ AWS Glue Data Quality - Automatic data quality rule recommendations based on your data AWS Glue for Ray - Data integration with Ray (ray.io), a popular new open-source compute framework that helps you scale Python workloads

https://ray.io (just dropping the link)

I've wondered whether it's easier to add data analyst stuff to Elixir that Python seems to have, or add features to Python that Erlang (and by extension Elixir) provides out of the box.

By what I can see, if you want multiprocessing on Python in an easier way (let's say running async), you have to use something like ray core[0], then if you want multiple machines you need redis(?). Elixir/Erlang supports this out of the box.

Explorer[1] is an interesting approach, where it uses Rust via Rustler (Elixir library to call Rust code) and uses Polars as its dataframe library. I think Rustler needs to be reworked for this usecase, as it can be slow to return data. I made initial improvements which drastically improves encoding (https://github.com/elixir-nx/explorer/pull/282 and https://github.com/elixir-nx/explorer/pull/286, tldr 20+ seconds down to 3).

[0] https://github.com/ray-project/ray

Adding these for anyone who had a similar question about Ray vs dask 1, 2, 3