Python Deep Learning Cookbook

How to do it...

1. Let's start with importing the libraries as follows:

import unidecodeimport stringimport randomimport mathimport torchimport torch.nn as nnfrom torch.autograd import Variable

2. As input and output, we can use any character:

all_characters = string.printableinput_size = len(all_characters)output_size = input_sizeprint(input_size)

3. We will be using a dataset with speeches from Obama:

filename = 'Data/obama.txt'data = unidecode.unidecode(open(filename).read())len_data = len(data)

4. We need to define the hyperparameters before moving on:

n_steps = 2000batch_size = 512hidden_size = 100n_layers = 2learning_rate = 0.01len_text = 200print_every = 50

5. Let's define a function that transforms characters to tensors:

def char_to_tensor(string): ...