To get the most out of this book

To fully understand all the algorithms in this book, it's important to have a basic knowledge of linear algebra, probability theory, and calculus.

All practical examples are written in Python and use the scikit-learn machine learning framework, Natural Language Toolkit (NLTK), Crab, langdetect, Spark (PySpark), Gensim, Keras, and TensorFlow (deep learning frameworks). These are available for Linux, macOS X, and Windows, with Python 2.7 and 3.3+. When a particular framework is employed for a specific task, detailed instructions and references will be provided. All the examples from chapters 1 to 14 can be executed using Python 2.7 (while TensorFlow requires Python 3.5+); however, I highly suggest using a Python 3.5+ distribution. The most common choice for data science and machine learning is Anaconda (https://www.anaconda.com/download/), which already contains all the most important packages.