This academic project focused on evaluating how different modeling approaches perform on high-dimensional sequence classification tasks.

Highlights

• Built a benchmarking framework for comparing Bi-LSTM models, PyTorch neural networks, and statistical baselines.
• Organized experiments so results could be compared more consistently across methods.
• Used the project to study tradeoffs between model complexity, accuracy, and efficiency.

The project strengthened my understanding of experimental design for machine learning systems and how to turn model evaluation into a reproducible engineering workflow.