We will not be implementing a full working neural network that performs the word embedding procedure, however we will be using a Python package called gensim to do this work for us:
# import the gensim package import gensim
A gensim can take in a corpora of text and run the preceding neural network structure for us and obtain word embeddings with only a few lines of code. To see this in action, let's import a standard corpus to get started. Let's set a logger in our notebook so that we can see the training process in a bit more detail:
import logging logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)
Now, let's create our corpus:
from gensim.models ...
