A list of resources for my friends at Pipedrive =)
- Awesome Machine Learning A curated and updated list of ML libraries, frameworks etc.
- Awesome Production Machine Learning A curated and updated list of Production ML resources.
- Awesome Industry Machine Learning
- Awesome Data Engineering A curated and updated list of Data Engineering resources.
- Machine Learning Examples A list of various ML code examples, with some relevent to reccomender systems. Linked to the Lazy Programmer courses on Udemy.
- Data Science Notebooks Huge list of Data Science notebooks.
- Google Rules of ML Google's rules of ML. More developer / production focused, but rules 1-7 are very important at the outset of an ML project.
- Google Good Data Analysis Excellent guide on how to approach data analysis.
- Google AI Software Engineering Research Publications List of Google produced papers relevent to AI & Software development. Very specific topics but lots of them.
- Machine Learning Yearning Book by Andrew Ng on technical strategy for ML + AI.
- Machine Learning Mastery Excellent resource site for ML, Statistics & Analysis. I try not to use too many random tutorials since they can be written by idiots, however this site is excellent!
- Chris Albon's site Another great non-idiot site focused more on Pandas and SKL.
- Google ML Glossary Gloassary of Machine Learning terms.
- R2D3 A visual introduction to Machine Learning. Very pretty!
- Dask Parallel processing for Python Analytics encompassing Numpy, Pandas and SKL.
- Modin Speed up Pandas with one line of code. Limited but easy.
- Dask on AWS with Fargate Setting up a Dask cluster on AWS using Fargate (Amazon's version of Kubernetes). Not tried this so may not work!!
- ScatterText Interactive sentiment visualisation.
- YellowBrick Machine Learning visualisation. Being developed quite quickly - everytime I check back there seems to be new things!
- FastText Facebook developed library for text classification. Very easy to use.
- Textract Python library that's brilliant for extract text from various documents (pdf, word, pictures etc.)
- VADER Lexicon based sentiment analysis library. Very quick and easy to use. May have been incorporated into NLTK.
- MLFlow Not had a chance to check this out properly yet, but I've heard very good things about it. Helps manage the Machine Learning lifecycle.
- LIME Again, not had a chance to check this out properly yet, but friends say it's good! Explainability framework for Python / R and can be used with SKL / Keras etc.
- DfT Data Science Team Processes The processes my team came up with @ DfT. This includes general stuff as well as a script to make Jupyter Notebooks easier to QA through
nbstripout
- The Institute for Ethical ML This is run by my friend Alejandro and has advice on how to develop ethical ML systems. Also he runs an ML Engineer newsletter that contains lots of cutting edge resources.
- Uber Ludwig A Deep Learning platform that allows training / testing of Deep Learning models without the need to write code.
- Netflix ML Research A lot of resources related to recommender systems, particularly the business side rather than the technical side.
- Netflix Recommender Research Specifc area of Netflix research for recommender systems.
- Recommending and Searching @ Spotify Presentation on how Spotify recommend and search.
- How Spotify’s Algorithm Knows Exactly What You Want to Listen To Explanation of Spotify's BeRT algoritm.
- Explore, Exploit, and Explain: Personalizing Explainable Recommendations with Bandits (Spotify) Paper by Spotify on the BeRT algorithm.