This is my solution to the Kaggle challenge Dogs vs. Cats.
In this competition, we have to write an algorithm to classify whether images contain either a dog or a cat.
The dataset provided by Kaggle contains 25,000 images of dogs and cats.
I used a neural network model, DenseNet, trained on ImageNet and available from torchvision.
I achieved an accuracy rate of 97% on new images.
The Jupyter Notebook is directly exported from Google Colaboratory.