Grad-Cam: Feature Importance for Convolution Neural Networks

At GTC-20 I saw a presentation about explainability for Convolution Neural Networks (CNN). I decided to create some code to apply it to video. You can see the results here.

Based on a presentation by Vineeth N Balasubramanian: https://www.iith.ac.in/~vineethnb/

Original Code:
https://keras.io/examples/vision/grad_cam/

GTC2020: https://www.nvidia.com/en-us/gtc/
My Github REPO for GTC2020: https://github.com/jeffheaton/present/tree/master/youtube/gtc-2020

** Follow Me on Social Media!
GitHub: https://github.com/jeffheaton
Twitter: https://twitter.com/jeffheaton
Instagram: https://www.instagram.com/jeffheatondotcom/
Discord: https://discord.gg/3bjthYv
Patreon: https://www.patreon.com/jeffheaton

Source of this machine learning/AI Video

AI video(s) you might be interested in …