Every quarter we ask the DagsHub team to round up their top libraries, machine learning articles and data science tips. Below is a list of our top five favorites covering topics on deep learning, explainability, data version control, and multi-tenancy issues in data science.

We are sure you will find this useful and if you have any of your own go-to resources feel free to share info@dagshub.com!

1. Should I Train a Model for Each Customer or Use One Model for All of My Customers?

This article has interesting links about how to decide whether to train a model per customer or one model for all customers. It highlights the advantages and disadvantages of four different strategies for dealing with the multi-tenancy problem in machine learning – It's well written and has personal insights into the process:

https://towardsdatascience.com/should-i-train-a-model-for-each-customer-or-use-one-model-for-all-of-my-customers-f9e8734d991

2. A Comprehensive Machine Learning Article: The ML and Deep Learning Compendium

A distillation of the knowledge of an experienced data scientist, this compendium aims to have every topic you might need in its table of contents, with concise explanations and links to further reading. With over 500 topics and a GitHub project, this compendium will save you countless hours googling and sifting through articles. Extremely practical and useful even for experienced data scientists who want to expand their knowledge.

https://book.mlcompendium.com/

3. Version Control with DVC in a nutshell

🥜(No Code!)

DVC is the most widely adopted open-source tool for large file versioning. The video goes into detail about DVC and covers the usage of DVC with Git to version both data and code. A must watch for anyone interested in approaches to data version control. Full disclosure: this video was produced by the DagsHub team and also gives examples of working with DVC and DagsHub.  

https://www.youtube.com/watch?v=QYNgWWearZ4

4. Awesome Explainable Graph Reasoning

What can possibly be better than Graph Machine Learning and GNNs? Well we think, this Explainable Graph Machine Learning Article! This is an excellent set of articles and software from AstraZeneca including a lot of great research on explainability and reasoning around graphs. It covers explainable predictions, explainable reasoning, code as well as some theory and survey papers

https://github.com/AstraZeneca/awesome-explainable-graph-reasoning

5. Interpretability and Explainability in Machine Learning

If you're looking for a machine learning article on explainability and interpretability, it might be hard to find good content, so we wanted to share a recommendation for the COMPSCI 282BR course by Hima Llakkaraju and Isaac Lage. With more deployments of models in finance, healthcare and even criminal justice it's essential that models are transparent and decision makers understand and trust predictions.

https://interpretable-ml-class.github.io/

Thats a wrap for this quarter. If you have any interesting data science or machine learning articles you would like to get featured and shared with our community please send us and email with your link and brief description to info@dagshub.com!