Model Evaluation

Hardik Chheda Updated by Hardik Chheda

After completing the model preparation process, Tellius displays the Evaluate Model page.

On the Evaluate Model page, you can view the details of the predictive analytics that Tellius performs by using the selected algorithms on the selected dataset.

You can also view different types of prediction errors in your predictive analytics.

The following list describes about the errors:

  • rmse (root mean squared error): This measure has the same units as the data series.
  • mse (mean squared error): This is used to refer to the unbiased estimate of error variance.
  • r2: R-Squared is a statistical measure of how close the data are to the fitted regression line.
  • Mae (mean absolute error): This gives the magnitude of the average absolute error. On the same page, you can also view the following tabs for prediction comparison with actual data.

The Actual Vs. Predicted tab displays the chart for the predicted values against the actual values.

The Regression plot tab displays the statistical details along with the residual error values in the prediction.

Click the Next button to save the model and also to use the model to predict another dataset. After the model evaluation is over, the Model Use page appears.

How did we do?

Model Selection

Model Use