Even though I tried my best to deliver the highest quality for the final version of the book, and I've received some great help from reviewers and editors, some (hopefully minor) errors still made it into the final product.
This page contains a list of known errors and related corrections.
If you find an error that is not listed here, please open a GitHub issue with a short description of the problem.
p. 40
The second example using the TweetTokenizer has a wrong reference to a non-existing TwitterTokenizer, and should be fixed as follows:
>>> from nltk.tokenize import TweetTokenizer
>>> tokenizer = TweetTokenizer()
p. 148
The script facebook_get_page_posts.py uses the hard-coded string 'PacktPub', instead it should
use the command line argument args.page as follows:
posts = graph.get_connections(args.page, 'posts', fields=all_fields)
Credits: Andre Logunov