Before Normalization, use the “Vector Assembler” node because Normalizer node takes the vectorudt datatype as input. (“VectorAssembler” node is used to assemble all features into vectors.)
In the configuration window, select the list of columns from the Input Columns box.
We have used the “Normalizer” node to normalize data.
In the configuration window, select Vectorudt into the Input Column Field and enter the desired name of the output column in Output Column Field.
On execution, output will show the normalized data as below.
Here, feature_v is input data and uninorm_feature_v is output column displaying the normalized data.
Hey Nagisa,
In Sparkflows, you can use the “Normalizer” node to normalize the data. Let us try to understand the process with the help of a simple workflow.
Below is the image of a simple workflow using the Normalizer node.
Before Normalization, use the “Vector Assembler” node because Normalizer node takes the vectorudt datatype as input. (“VectorAssembler” node is used to assemble all features into vectors.)
In the configuration window, select the list of columns from the Input Columns box.
We have used the “Normalizer” node to normalize data.
In the configuration window, select Vectorudt into the Input Column Field and enter the desired name of the output column in Output Column Field.
On execution, output will show the normalized data as below.
Here, feature_v is input data and uninorm_feature_v is output column displaying the normalized data.