This project consist of a Binary classifier based on the cats vs dogs dataset used in the following paper.
Elson, Jeremy and Douceur, John (JD) and Howell, Jon and Saul, Jared. (2007). Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization. Retrieved from https://www.microsoft.com/en-us/research/publication/asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization/
The project itself is served by a Flask application and has a training accuracy of 98% as well as a validation accuracy of 92%.
Feel free to download the project and try the application.
In order to run the project you need to install Tensorflow 2.2.0, Flask, and other basic libraries like Numpy. I highly recommend to use conda environments.
