Reseña
As the second edition of Python Data Science Essentials, this book offers updated and expanded content. Based on the recent Jupyter notebooks (based on interchangeable kernels, a truly polyglot data science system), this book incorporates all the main recent improvements in Numpy, pandas, and Scikit-learn. Additionally, it offers new content on deep learning (by presenting Keras - based on both Theano and Tensorflow), on beautiful visualizations (seaborn and ggplot), and on web deployment (using bottle).
This book starts by explaining how to set up your essential data science toolbox in Python's latest version, 3.5, using a single source approach (implying that the code in this book will be easily reusable in Python 2.7 as well). Then, it will guide you through all the data munging and preprocessing phases.
Finally, it will complete the overview by presenting you with the principal machine learning algorithms, graph analysis technicalities, and visualization and deployment instruments.