Generating bedroom photos from the LSUN dataset

LSUN (Large-scale Scene Understanding, https://www.yf.io/p/lsun) is a large image dataset with 10 scene categories and 20 object categories. You can get the downloading toolkit from https://github.com/fyu/lsun. We will use the bedroom category to train our DCGAN, which has more than 3 million bedroom photos:

$ git clone https://github.com/fyu/lsun.git$ cd lsun$ python download.py -c bedroom
You can also export the images as individual files with python data.py export bedroom_train_lmdb --out_dir bedroom_train_img so that you can easily use these images for other projects. But try not to directly open the image folder with your file manager. It will take a lot of RAM and time.

The dataset is ...

Get Hands-On Generative Adversarial Networks with PyTorch 1.x now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.