Design matrix

The whole dataset as one matrix -- one row per sample, one column per feature.

A matrix $\mathbf{X}$ of shape $(n, p)$ collecting $n$ data samples as rows and $p$ features as columns, so that row $i$ is the feature vector for sample $i$ and every model in this course is fit against this one shared representation.

Examples

A dataset of 500 houses with 8 features each is a $500 \times 8$ design matrix.
The target vector $\mathbf{y}$ has one entry per row of $\mathbf{X}$ .