Model Features

This section provides the option to select features that will be used for training, options to change the data type and transform each selected feature. Note that the target feature is not considered a training feature.

• In this section, the user can see the list of features with prefilled default values.
• The type of feature can be selected as date, index, categorical, numerical, or text.
• The fill method is supported with the default value for numerical being Median and for Categorical is Mode.
• Users can select different Encoding techniques for different features.
• Outlier and Normalization can be selected from the list of given options.
• Finally, mention the Target Features (for supervised learning). For some of the algorithms like Document similarity and clustering, the target feature is not applicable.