For the demo we used JupyterLab systems to run everything, the installation instructions can be found here:
https://jupyterlab.readthedocs.io/en/stable/
Once you have JupyterLab set up, you will need to navigate to the Demo folder from the main branch. In this folder you will be able to retrieve the pre-trained models, labeled "SingleModelPaper" and "ConvModelPaper", as well as the data set for the project, it is the folder labeled "miscImgs".
Please ensure your image folder "miscImgs" and the pre-trained models are in the same directory as your Python notebook.
Table of Contents
- Imports and general functions
2a. Single Layer Network
2b. Usefulness of Single Layer Network
3a. Convolution Network
3b. Usefulness of Convolution Network
- How to visualize the weights