onuricen / DoggoCNN

A Convolutional Neural Network capable of doing binary classification

DoggoCNN

DoggoCNN is a Convolutional Neural Network capable of classifying cats and dogs with %81 accuracy.

• Trained on 4000 cat and 4000 dog pictures (Small version of this Kaggle Dataset) )
• %81 accuracy on training set

Tech

DoggoCNN uses a number of open source projects to work properly:

• Keras
• Cv2
• NumPy

Usage

Use predict.py to do predictions (You can use default jpg files that comes with this repo)

$ predict.py pup.jpg

Training

You can train the model to do any kind of binary classification.For this example I've used dogs and cats dataset because of convenience

Use cnn.py to train your own model

Licence

This project is available as open source under the terms of the MIT License.