WhiskerWoof

Hey There 👋. This here's an implementation of a convolutional neural network (CNN) built with PyTorch for classifying images of cats and dogs. The model runs on test data with 76.01% accuracy

You can download the dataset at https://www.microsoft.com/en-us/download/details.aspx?id=54765

Before you start

In your project root directory, make sure to create a 'data' folder, and each image class to have a subfolder inside it.

The main.py file contins code for testing my pre-trained WhiskerWoof model. Feel free to tinker around and train further

Note: Some images in the dataset are truncated and/or empty and may have to be deleted. I'd rather you execute a python script than delete manually

You'll Need

• Python 3.8 or higher
• torch
• matplotlib
• torchvision
• scikit-learn
• Pillow

Additional required packages can be found in the requirements.txt