Putting CGAN to work on some examples
Now that the CGAN
class is completed, let's go through some examples in order to provide you with fresh ideas on how to use this project. First of all, we will have to get everything ready for both downloading the necessary data and training our GAN. We start by importing the routine libraries:
import numpy as np import urllib.request import tarfile import os import zipfile import gzip import os from glob import glob from tqdm import tqdm
We then proceed by loading in the dataset and CGAN
classes that we previously prepared:
from cGAN import Dataset, CGAN
The class TqdmUpTo
is just a tqdm
wrapper that enables the use of the progress display also for downloads. The class has been taken directly from the project's page at https://github.com/tqdm/tqdm:
class TqdmUpTo(tqdm): """ Provides `update_to(n)` which uses `tqdm.update(delta_n)`. Inspired by https://github.com/pypa/twine/pull/242 https://github.com/pypa/twine/commit/42e55e06 """ ...