GitHub is a great place to learn new things, find something useful for current projects, and get inspired for future projects. This list of cool projects is just a small part of the variety of interesting ML repositories that you can find on github.
Python is the language that most people write in machine learning and data science. And it’s a beautiful language – simple, readable, with its own PEP8 standard. But this language has a significant drawback – its speed. Therefore, if you need your ML project to be fast and, at the same time, written in Python, then the XLearn library is created especially for you. It has C ++ under the hood and, according to the developers, can increase the speed of the code by 5-13 times compared to similar Python ML libraries.
This is a fairly well-known utility created by Facebook for automatic time series prediction. The best part is that Prophet can handle gaps in the data with ease. Also – this is an open-source project, with the support of the IT giant, which gives some freedom for creativity.
Another library for working with rows. Seqlearn is easy to use – it is based on Numpy and Scikit-Learn.
MLBox is a very powerful library for automated machine learning. It can be useful in data reading, feature selection and feature engineering, as well as in the selection of model parameters.
It is an open source library that can help you develop ML applications. With the help of ModAL, you can make your code more modular: you don’t have to rewrite the same thing several times. Also, ModAL has many ready-made functions, which also simplifies development.