In the following few lectures you will learn to train word vectors using some text corpora and to use pre-trained word vectors for text processing tasks such as classification etc.
Let’s train the word vectors on a larger, more realistic dataset now- the text8 corpus, which is a compressed version of Wikipedia.
Let’s now visualize the word vectors obtained earlier. Since our word vectors are 100- dimensional, we will not be able to visualize them in the 100-dimenional space. thus, we’ll reduce them to a two-dimensional space using PCA and plot them on a 2-D plane