The observed responses are subtracted from the projected responses to yield residuals, which are estimates of experimental error. After all of the unknown model parameters have been determined from the experimental data, the anticipated response is calculated using the chosen model.
The 'delta' between the actual target value and the fitted value is the residual in machine learning. In regression issues, residual is a significant notion. It is the foundation of all regression metrics, including mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE).