Assignment 3

For this lab, we recommend that students watch the video and read the material in the references section of this page.

In this assignment, we will explore adversarial generative networks (GAN). The objective is to see in practice the concepts covered in the classroom. We provide two example notebooks; the first addresses the implementation of a GAN using Multilayer Perceptron, and the second shows the implementation of a GAN using convolutional neural networks. Check out the notebooks below.

1. Implementing a GAN using Multilayer Perceptron, click here.
2. Implementing a GAN using convolution networks, click here.

Suggestions:

1. Create a copy of the notebook.

2. Run the cells; we recommend running them one by one to get familiar with the process.
3. Make modifications to the training block. You can change the batch size, LeakyReLU slope value, and training rate. Compare the result with the model of the example notebooks.

References

Online Course 2021- Image Processing