/Image-Generation-using-GANS

Using MNIST data I trained my GAN network to replicate images of digit 5 in a way as diluted as possible.

Primary LanguageJupyter Notebook

Image-Generation-using-GANS

Machine learning Concepts Used

  • Generative adversarial Networks
  • Forward Propagation
  • Backward Propagation
  • Loss function of Generator and Discriminator

ML libraries used

  • Tensorflow
  • Matplotlib
  • PIL
  • numpy
  • keras

I have trained a simple GAN model to generate digit 5. The dataset used in of MNIST dataset which is included in the keras library.

Results

image_0

This image was generated using generator model of GAN

Check out my notebook where i have discussed the details of dealing with the this kind of neural network. Also it can take a lot of time to run on cpu so it adivised to run it on kaggle gpu. I have uploaded the notebbok there as well. Kaggle notebook