A 6-part series of hands-on Jupyter notebooks covering advanced data science topics including unsupervised learning, probabilistic modeling, neural networks, CNNs, sequence models, and transformers. The sequel to Foundations of Data Science.
Overview
This series picks up where Foundations of Data Science leaves off, diving into advanced topics across six progressive notebooks. Starting with unsupervised learning (clustering, PCA), it moves through Bayesian probabilistic modeling, neural networks from scratch, convolutional neural networks for image classification, sequence models and NLP, and culminates with attention mechanisms and transformers. Each notebook includes original simulated datasets, exercises with solutions, and practical implementations.
Who This Is For
Students and practitioners who have completed Foundations of Data Science or have equivalent knowledge of Python, statistics, and supervised learning
What's Included
- 6 progressive notebooks building from unsupervised learning to transformers
- Unsupervised Learning — K-means, hierarchical clustering, DBSCAN, PCA
- Probabilistic Modeling — Bayesian inference, PyMC, hierarchical models, MCMC
- Neural Networks — backpropagation from scratch, Keras, regularization, bootstrap prediction intervals
- CNNs — convolution, pooling, Fashion-MNIST, data augmentation, transfer learning (MobileNetV2)
- Sequence Models & NLP — RNN, LSTM, GRU, bidirectional models, text generation
- Attention & Transformers — multi-head attention, positional encoding, Vision Transformer on CIFAR-10