Matrix Multiplication From Scratch

Difficulty: Medium · Concept: Linear Algebra — Matrices

For an $(m \times n)$ matrix $A$ and an $(n \times p)$ matrix $B$, the product $C = AB$ is $(m \times p)$, where each entry $C_{ij} = \sum_k A_{ik} B_{kj}$ — the dot product of $A$'s row $i$ with $B$'s column $j$. This is the single operation every linear layer in a neural network runs on every forward pass.

Your task: implement matmul(A, B) for matrices represented as lists of lists of numbers (no NumPy). Raise a ValueError if the inner dimensions don't match (A's number of columns must equal B's number of rows).

Next: Mean Squared Error From Scratch