fish_classification - This project involves using CNNs and PyTorch to classify fish images into different categories. We will train a model on a dataset of fish images and use it to predict the class of new, unseen images. Finally its performance is compared to a fine-tuned AlexNet and a MobileNet to explore Transfer Learning.
object_detection - A Faster R-CNN is trained to detect and categorize maritime objects, such as sailboats and buoys, at sea. Each detection is visualised with a bounding box and relevant metrics are displayed next to the image category. For this purpose images are filtered from the Pascal VOC for training.
mnist_class- This project involves using MLPs and PyTorch to classify handwritten letters from the MNIST dataset.