Create composite transformer spark

Question

I am using an NGram Transformer then a CountVectorizerModel.

I need to be able to create a composite transformer for reuse later.

I was able to achieve this by making a List<Transformer> and looping through all elements but I want to know if it is possible to create a Transformer using 2 other Transformer

Answer 1

This is practically very easy, you'll just need to use the Pipeline API to create your pipeline :

import java.util.Arrays;

import org.apache.spark.ml.Pipeline;
import org.apache.spark.ml.PipelineModel;
import org.apache.spark.ml.PipelineStage;
import org.apache.spark.ml.feature.CountVectorizer;
import org.apache.spark.ml.feature.NGram;
import org.apache.spark.ml.feature.Tokenizer;
import org.apache.spark.sql.Row;
import org.apache.spark.sql.RowFactory;
import org.apache.spark.sql.types.DataTypes;
import org.apache.spark.sql.types.Metadata;
import org.apache.spark.sql.types.StructField;
import org.apache.spark.sql.types.StructType;

List<Row> data = Arrays.asList(
            RowFactory.create(0, "Hi I heard about Spark"),
            RowFactory.create(1, "I wish Java could use case classes"),
            RowFactory.create(2, "Logistic,regression,models,are,neat")
    );

StructType schema = new StructType(new StructField[]{
            new StructField("id", DataTypes.IntegerType, false, Metadata.empty()),
            new StructField("sentence", DataTypes.StringType, false, Metadata.empty())
});

Now let's define our pipeline (tokenizer, ngram transformer and count vectorizer) :

Tokenizer tokenizer = new Tokenizer().setInputCol("text").setOutputCol("words");

NGram ngramTransformer = NGram().setN(2).setInputCol("words").setOutputCol("ngrams");

CountVectorizer countVectorizer = new CountVectorizer()
  .setInputCol("ngrams")
  .setOutputCol("feature")
  .setVocabSize(3)
  .setMinDF(2);

We can now create the pipeline and train it :

Pipeline pipeline = new Pipeline()
            .setStages(new PipelineStage[]{tokenizer, ngramTransformer, countVectorizer});

// Fit the pipeline to training documents.
PipelineModel model = pipeline.fit(sentenceDataFrame);

I hope this helps