This repository contains information and documentation for working with the 'Computer' part of the AmazonCoBuy dataset. This dataset is part of the larger AmazonCoBuy collection and focuses on co-purchasing behavior among computer-related goods.

• Nodes: 13,752 (represent goods)
• Edges: 491,722 (indicate that two goods are frequently bought together)
• Node Features: 767 (bag-of-words encoded product reviews)
• Class Labels: 10 (product categories)
• Self-Loops: Self-loops are added to the graph regardless of whether they already exist

• Implemented the Graph Convolutional Network (GCN) model with 2 layers and trained it using the Adam optimizer with cross-entropy selected as the loss function.
• Achieved best validation accuracy of 87.3%.
• Python libraries used: DGL, PyTorch, Matplotlib, and Seaborn.

This dataset and preprocessing information can be used for tasks such as node classification, link prediction, and other graph-based machine-learning tasks. Ensure that the necessary preprocessing steps, such as adding self-loops and defining train/validation/test masks, are completed before training models on this dataset.