PyTorch Loader for the CelebA dataset with the identities of the people in the images as labels.
Run downlad.sh: bash download.sh celeba (this script is from https://github.com/yunjey/stargan.git)
now, place the identeties (identity_CelebA.txt) next to the images. The directories would look like: stargan/data/celeba: "identity_CelebA.txt images list_attr_celeba.txt". You can also find the identities here: https://drive.google.com/drive/folders/0B7EVK8r0v71pWEZsZE9oNnFzTm8 .
now, run the test_celeba-with-id.ipynb notebook to see how you can also get the ids. In case of question, email me at [email protected].