Learns the mean of its input data.
A data set.
A mean learning algorithm.
A trained regressor. Signal Predictor sends the regressor only if signal Data is present.
This is the simplest learner widget for regression problems. It learns the mean of the class variable and returns a predictor with the same mean value. Due to its accuracy, this widget can serve as a baseline for other regression models.
- Learner/predictor name
- Produce a report.
- The Apply button commits changes to the output. Alternatively, tick the box on the left side of the button to apply changes automatically.
In the first example, we use Mean Learner to construct a predictor and input it into the Data Table. We used the housing data set. In the table, you can see an extra column Mean Learner with one (mean) value for all instances.
Another way to use Mean Learner is to compare it to other learners in the Test&Score widget.