Classifies whether an image is of a dog or cat based on the features
The project is about classifying images of cats and dogs. I have used pretrained model densenet121 which is trained on Imagenet dataset.
So, that model will be a good feature extractor and I am changing the classifier part to classify the images of cats and dogs.
After training for just one epoch the model gave and accuracy of over 98%.
To the same model to classify the images of cats and dogs, you don't need to train again. I have uploaded checkpoint.pth
file that contains, number of input, hidden and output nodes and also the state_dict. You build the same classifier and load the
state dict into the model and directly use that.
- Python: language
- NumPy: library for numerical calculations
- Matplotlib: library for data visualisation
- Pytorch: a deep learning framework by Facebook AI Research Team for building neural networks
- torchvision: package consists of popular datasets, model architectures, and common image transformations for computer vision
To use this project, clone the repo
git clone https://github.com/Surya-Prakash-Reddy/Classifying-Cats-and-Dogs.git
After cloning, you can use the Classifier.ipynb notebook to learn or modify. You can also use checkpoint.pth file if you do not wish to train your model again or build some webapps using the model. If you are using checkpoint.pth, you can find the architecture in Classifier.ipynb