AI / Machine Learning

AI Intro Course Labs

A series of machine learning labs completed in Google Colab, building from classical algorithms through neural networks on real-world datasets.

Methods

• Naive Bayes, KNN, K-Means clustering
• MLP and CNN architectures
• Convolutional auto-encoders

Key assignments

• Celestial object classification on Sloan Digital Sky Survey data.
• Spotify song clustering by audio features.
• Anime rating prediction with mixed feature types (MLP).
• Pokemon image classification — 0.945 validation accuracy; auto-encoder experiments.

Takeaway

These labs gave me hands-on experience with the full ML pipeline — data loading, model design, training, and evaluation — using NumPy, PyTorch, and Keras.