The project is about Generative Adversarial Networks(GANs) paper to generate new images of faces. I utitlized the CelebFaces Attributes Dataset(CelebA) to train the model. CelebA is a large-scale face attributes dataset with more than 200K celebrity images, each with 40 attribute annotations. The images in this dataset cover large pose variations and background clutter.
View samples of images from the generator.Gradually,the Generator learned to create new images.
- Udacity Deeplearning ND
- DCGAN Tutorial from PyTorch
- Generative Models - openai blog
- How to Train a GAN? Tips and tricks to make GANs work？- Soumith Chintala，Emily Denton，Martin Arjovsky，Michael Mathieu
- Generative Adversarial Nets
- Batch Normalization
- Improved Technigques for Training GANs
- Image-to-Image Translation with Conditional Adversarial Networks
- StackGAN
- CycleGAN and pix2pix
- Generative models
- conda 4.7.12
- pytorch 1.3.0
- python 3.7.5
We use SemVer for versioning. For the versions available, see the tags on this repository.
- Pytorch - An open source machine learning framework that accelerates the path from research prototyping to production deployment.
- Tom Ge - Fullstack egineer - github profile
This project is licensed under the MIT License