Show HN: MLEM – ML model deployment tool | Hacker News Reader

Show HN: MLEM – ML model deployment tool

24 pointsaguschin11 comments

Hi, I'm one of the project creators. MLEM is a tool that helps you deploy your ML models. It’s a Python library + Command line tool.

1. MLEM can package an ML model into a Docker image or a Python package, and deploy it to, for example, Heroku.

2. MLEM saves all model metadata to a human-readable text file: Python environment, model methods, model input & output data schema and more.

3. MLEM helps you turn your Git repository into a Model Registry with features like ML model lifecycle management.

Our philosophy is that MLOps tools should be built using the Unix approach - each tool solves a single problem, but solves it very well. MLEM was designed to work hands on hands with Git - it saves all model metadata to a human-readable text files and Git becomes a source of truth for ML models. Model weights file can be stored in the cloud storage using a Data Version Control tool or such - independently of MLEM.

Please check out the project: https://github.com/iterative/mlem and the website: https://mlem.ai

I’d love to hear your feedback!

Awesome project! Congratulations on Shipping! What ML frameworks are you going to support?

Thank you! MLEM already supports Scikit-learn, PyTorch, XGBoost, LightGBM and CatBoost now. We're going to add fastai, TensorFlow and Huggingface very soon, and then there are many others we'd like to support! Feel free to post an issue in the GH repo for the framework of your choice)

Sure, thank you!

Oh, and MLEM can also serialize any Python function. (How could I forget!)

This looks pretty cool! I'm gonna keep my eyes on this for more deployment options. The world needs a (good) open-source swiss-army-knife for ML serving

Thanks! We're going to add AWS Sagemaker, pure Kubernetes and Seldon-core soon. Feel free to drop an issue in GH if you need something in particular!

It would be helpful to see the difference with MLFlow. The deployment part as well as model registry part.

Hi! Thanks for the question! There are a few important differences:

- MLEM automatically extracts the metadata from the model for you. With MLflow, you need to specify ML framework and environment.

- For the Model Registry that you can build in Git with MLEM, you don't need a separate service and Database up, except for GitHub or GitLab.

can't wait to put "mlem experience required" in a job posting and stare intently at my HR

Yep! Machine Learning should be mlemming :)

+1

Reply on news.ycombinator.com