The project is inspired by the work of Jason Brownlee here
The goal is to create both a generation model (generator) and a model to identify fakes (discriminator). Both will learn from each other
During exploration, the point will be to create both models separatly.
- Discrimator Model (exploration & script)
- Generator Model (exploration & script)
After the creation of both models, the aim will be to regroup all methods inside the same script for a generating MNIST handwritting digits.
- GAN script
The final step is to see if it's possible to generate handwritting text (if the data exist, we can do it)