Practice Notebooks
Work through each notebook sequentially. Complete the exercises to unlock the next one.
0/3
1
Convolutions from Scratch
Implement the convolution operation from scratch, understand learnable filters, ReLU, and max pooling, then build and train a complete CNN on CIFAR-10.
50 min2 exercisesNarrated
Vision Transformers from Scratch
Build the Vision Transformer from first principles: patch embeddings, self-attention, and the full encoder — all implemented manually before using PyTorch.
65 min2 exercisesNarrated
CNN vs. Vision Transformer: A Hands-On Comparison
Train both a CNN and a ViT on the same dataset and directly compare their behavior, accuracy, computational cost, and what they learn.
55 min2 exercisesNarrated