Free interactive textbook combining mathematical explanations, code, and practical deep-learning concepts.
Authored by Aston Zhang, Zachary C. Lipton, Mu Li, and Alexander J. Smola, Dive into Deep Learning (D2L) is an interactive open-source textbook adopted by hundreds of universities worldwide. Every section pairs mathematical derivations with runnable PyTorch code—bridging linear regression and softmax classification in Phase 2 directly to multilayer perceptrons, initialization, normalization, and modern deep architectures in Phase 3.
Course status is tracked independently from roadmap topic checkboxes so you can mark external courses without altering your topic mastery percentage.
Cross-Phase Interactive Textbook (Phase 2 Module 2.2 & Phase 3 Modules 3.1–3.4)
Entire interactive textbook and runnable PyTorch notebooks are 100% free and open-access online.
Status: Verified (2026-11-09)
Review these foundations before starting so you get the most out of this material:
In Phase 2: Machine Learning From Scratch
Read Chapters 3 and 4 (Linear Neural Networks for Regression and Classification) to see how classical linear and softmax regression transition into deep learning.
In Phase 3: Deep Learning & Neural Networks
Study Chapters 5, 6, and 12 (Multilayer Perceptrons, Forward/Backward Propagation, Numerical Stability/Initialization, and Optimization Algorithms) alongside Phase 3.
Mathematical formulation and NumPy implementation of linear regression, logistic regression, MSE, binary cross-entropy, and decision boundaries.
Classification and regression metrics, confusion matrices, precision-recall tradeoffs, ROC-AUC, baseline comparisons, and leakage detection.
Neurons, feedforward layers, modern activations (ReLU, GELU, SiLU), forward/backward passes, initialization, normalization, and residual streams.
PyTorch tensors, GPU acceleration, autograd mechanics, nn.Module architecture, DataLoaders, checkpointing, and mixed precision.
Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), LSTMs, Autoencoders, Generative Models, and Reinforcement Learning fundamentals.
A code-first introduction to neural networks covering automatic differentiation, backpropagation, language models, and GPT implementations.
Official 'Learn the Basics' tutorial series covering PyTorch Tensors, Datasets/DataLoaders, `nn.Module`, `torch.autograd`, optimization loops, and model checkpointing.
Mathematical foundations of machine learning, including linear algebra, vector calculus, probability, and optimization.