ibm-research/AIX360-CEM-MAF-USECASE

Models for usecase with AIX360 CEM-MAF

Refer to notebook at CEM-MAF example for further information. The notebook automatically downloads all models contained here.

attr_model/*: These are attribute models used to describe images.

gan_model/* Contains the GAN model used to generate new images.

celebA: 8-class classifier learned from the following binary attributes: Young/Old, Male/Female, Smiling/ NOt Smiling. This is trained from a Resnet50 [1] architecture to classify images from the CelebA [2] dataset of celebrity faces.

[1] Shaoqing Ren, Kaiming He, Xiangyu Zhang, and Jian Sun. Deep Residual Learning for Image Recognition. CVPR (2016).

[2] Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. Deep Learning Face Attributes in the Wild. ICCV (2015).

Note regarding GAN model:

The GAN model is provided for reference purposes only and is subject to different licensing: Creative Commons Attribution-NonCommercial 4.0 International Public License. The original model is a file karras2018iclr-celebahq-1024x1024.pkl taken from the github project Progressive Growing of GANS.