We did not cover every machine learning package available for Python. Given the limited space, we chose to focus on scikit-learn. However, there are other options and we list a few of them here:
- pandas (https://pandas.pydata.org) – If you decide to fall in love with just one Python package in your whole life, choose this one! It provides a convenience layer on top of NumPy and speeds up common tasks, such as interactive data preprocessing, tremendously.
- Of course, all the other exciting deep learning toolkits, such as CNTK (http://cntk.ai), PyTorch (https://pytorch.org/), MXNet (https://mxnet.apache.org/), Chainer (https://chainer.org/), DSSTNE (https://github.com/amzn/amazon-dsstne), or DyNet (https://github.com/clab/dynet).
- Keras (https://keras.io/), which is a convenient library on top of TensorFlow and CNTK. Often, people start with a "real quick trying-out" version of Keras, only to find out that it is already good enough.
- MissingNo (https://github.com/ResidentMario/missingno), which is a Python module dedicated to analyzing and pruning missing values.
- Machine Learning Toolkit (Milk) (http://luispedro.org/software/milk) – This package was developed by one of the authors of this book and covers some algorithms and techniques that are not included in scikit-learn.