clear_ml::loss_functions

Function gradient_mse

Source 
pub fn gradient_mse(
    x: &Vec<Vec<f64>>,
    y_hat: &Vec<f64>,
    y: &Vec<f64>,
) -> Result<Vec<f64>, Error>
Expand description

Calculates the gradient of the mean squared error between two vectors.

§Formula

dMSE/dw = 2/n * sum((y_hat - y) * x)

§Arguments

  • x - Vector of input values
  • y_hat - Vector of predicted values
  • y - Vector of actual values

§Returns

  • Vector of partial derivatives of MSE with respect to each weight
  • The last element of the vector is the partial derivative with respect to the intercept