Project 2: Predicting Similar Food Taste

A neural network was trained to predict similar food tastes based on triplets of food pictures (anchor, similar taste, and distinct taste). Image embeddings were generated using ResNet50 by transforming, resizing, and normalizing the images. The neural network was designed with ReLU activation functions using the PyTorch library. The data was split into training and validation sets, and multiple epochs were used for training.

This Project’s GitHub Repository (currently not available)