mar 
How to tell a model that has learned from one that has memorised: the training/validation/test split, loss curves, a NumPy example and the usual defences (L1 and L2 regularisation, dropout and early stopping).
A decision guide to choosing between linear models, logistic regression, decision trees, random forests, gradient boosting and neural networks on tabular data, with NumPy code and the limits of each approach.
What changes inside a neural network when it “learns”: neurons, layers, the loss function, backpropagation and gradient descent, with a minimal NumPy version and the same loop in PyTorch.