Vector Indexer Transform

Vector Indexer indexes categorical features inside of a Vector. It decides which features are categorical and converts them to category indices. The decision is based on the number of distinct values of a feature.

Input

It takes in a DataFrame and transforms it to another DataFrame

Output

It indexes categorical features in datasets of Vectors and stores the result into a new column of the DataFrame.

Type

ml-predict

Class

fire.nodes.etl.NodeVectorIndexerTransform

Fields

Details

Vector Indexer Transform Node Details

The Vector Indexer Transform Node is used to index categorical features inside of a Vector. It takes in an input DataFrame and transforms it to another DataFrame. The decision of which features are categorical is based on the number of distinct values of a feature. The transformed DataFrame contains a new column with the indexed categorical features.

Input Parameters

DataFrame : The input DataFrame which contains the features to be indexed.

Examples

Vector Indexer Transform Node Example

Consider the following example, where we have a DataFrame with a column ‘features’ containing Vector values. We use the Vector Indexer Transform Node to index the categorical features in the ‘features’ column.

Input DataFrame:

id features

1 [1.0,2.0,3.0,”dog”]

2 [4.0,5.0,”cat”,6.0]

3 [7.0,8.0,”dog”,9.0]

4 [10.0,”cat”,11.0,12.0]

Output

id features indexed_features

1 [1.0,2.0,3.0,”dog”] [1.0,2.0,3.0,0]

2 [4.0,5.0,”cat”,6.0] [4.0,5.0,1,6.0]

3 [7.0,8.0,”dog”,9.0] [7.0,8.0,0,9.0]

4 [10.0,”cat”,11.0,12.0] [10.0,1,11.0,12.0]

In this example, the input column is ‘features’ and the output column is ‘indexed_features’. The Vector Indexer Transform Node indexes the categorical features in the ‘features’ column, creating a new column ‘indexed_features’ with the indexed categorical features.