Deep Learning Theory and Practice covers: Hands-On Machine Learning, Deep Learning, Machine Learning with PyTorch and Scikit-Learn. Understand neural networks, backpropagation, CNNs, RNNs, transformers, and build deep learning models. This is where you go from classical ML to the technology behind modern AI. If classical ML (Module 3) was like learning Rails, deep learning is like learning how Rails itself works under the hood — Rack, middleware, the router, ActiveRecord internals. You go from "I can build apps" to "I understand the framework deeply enough to extend it." Same jump here: from "I can use ML models" to "I can build and modify the models themselves."
This course unlocks once you've finished its prerequisite. Open prerequisite →
This is the module where everything changes. Not because deep learning is inherently harder than classical ML — in some ways it's simpler (fewer hyperparameters to tune, less feat…
The book you're opening is the same one sitting on your desk from Course 4 — and that's the point. Chapter 9 ended with you clustering data using algorithms a statistician from 19…
There is exactly one book in this path that researchers cite the way lawyers cite precedent, and you're holding it. When a paper says "we use the standard formulation (Goodfellow …
There used to be a framework debate in deep learning. There isn't anymore: research runs on PyTorch, the models you'll build in Course 7 are PyTorch, and the from-scratch LLM book…
Work through each item before the checkpoint.
Build a convolutional network in PyTorch with no pretrained weights — convolutions, pooling, and fully-connected layers wired by hand — and train it until it actually learns. The …
The other half of deep learning: a model that processes text. Build a character-level sequence model in PyTorch from scratch — tokenizer, embeddings, and either a recurrent cell o…
7 lessons. Read in order; spiral back when you need to. By the end you'll have used the core ideas twice — once on the abstract, once on something you'll meet at work next week.