Dataset contains about 60k training images and 20k validation images.
DID NOT INCLUDE MODEL WEIGHTS IN REPO
1. Image and classification output works- Livefeed with very high latency.
- Algorithm runs into memory error after listening for long enough.
- Fix problems