Features are the basic building blocks of datasets. The quality of the features in your dataset affects the quality of the insights you will gain when you use that dataset for training and evaluating models.
Predictions (or targets) are the variables used to train the model or evaluate the accuracy and precision, in combination with the rest of the dataset features. For training and evaluation purposes, the features and targets data must be historical and curated.
Property | Description |
---|---|
Name | Specifies the name of the variable. |
Group | Specifies the group the variable belongs to. |
Type | Specifies the data type of variable as Continuous or Categorical. |
Importance | (Applicable for features) Measures the increase in the prediction error of the model after the feature's values are permuted, which breaks the relationship between the feature and the true outcome. The importance of a feature is measured by calculating the increase in the model's prediction error after permuting the feature.
|
PSI drift value | Specifies change in the distribution of a variable over time to detect model performance issues. |