Making predictions with matrix multiplication

In later chapters, you will learn to train linear regression models. This process will yield a vector of parameters that can be multiplied by the input data to generate a vector of predictions. In the exercise, we will use the following tensors:

\(X = \begin{bmatrix} 1 & 2 \\ 2 & 1 \\ 5 & 8 \\ 6 & 10 \end{bmatrix}\), \(b = \begin{bmatrix} 1 \\ 2 \end{bmatrix}\), \(y = \begin{bmatrix} 6 \\ 4 \\ 20 \\ 23 \end{bmatrix}\)

X is the matrix of input data, b is the parameter vector, and y is the target vector. You will use matmul() to perform matrix multiplication of X by b to generate predictions, ypred, which you will compare with y. Note that we have imported matmul() and constant().

Define X, b, and y as constants.
Compute the predicted value vector, ypred, by multiplying the input data, X, by the parameters, b. Use matrix multiplication, rather than the element-wise product.
Define error as the targets, y, minus the predicted values, ypred.

Take Hint (-30 XP)

Exercise

Making predictions with matrix multiplication

Instructions