This project was I wanted to create a classification model that could classify images of domino tiles. My long term goal is to create a visual tool to calculate points on a domino game.
The data was created by me and consists of 28 classes (folders) of domino tiles with 100 images in each (100x100px). The data can also be downloaded through Kaggle: https://www.kaggle.com/bjorkwall/photographs-of-28-different-domino-tiles
The project consists of the data (photos) with labeled folders to be used as target features as well as a Jupyter Notebook with the model. The accuracy is about 99.2%.
Required installs can be found in requirements.txt (created using 'conda list --export').