1. Learn
    2. /
  2. Courses
    3. /
  3. Recurrent Neural Networks (RNN) for Language Modeling in Python

Exercise

Preparing text data for model input

Previously, you learned how to create dictionaries of indexes to words and vice versa. In this exercise, you will split the text by characters and continue to prepare the data for supervised learning.

Splitting the texts into characters may seem strange, but it is often done for text generation. Also, the process to prepare the data is the same, the only change is how to split the texts.

You will create the training data containing a list of fixed-length texts and their labels, which are the corresponding next characters.

You will continue to use the dataset containing quotes from Sheldon (The Big Bang Theory), available in the sheldon_quotes variable.

The print_examples() function print the pairs so you can see how the data was transformed. Use help() for details.

Instructions

100 XP
  • Define step equal to 2 and chars_window equal to 10.
  • Append the next sentence to the variable sentences.
  • Append the correct position of the text sheldon to the variable next_chars.
  • Use the print_examples() function to print 10 sentences and next characters.