Core Spark functionality.
Core Spark functionality. org.apache.spark.SparkContext serves as the main entry point to Spark, while org.apache.spark.rdd.RDD is the data type representing a distributed collection, and provides most parallel operations.
In addition, org.apache.spark.rdd.PairRDDFunctions contains operations available only on RDDs
of key-value pairs, such as groupByKey and join; org.apache.spark.rdd.DoubleRDDFunctions
contains operations available only on RDDs of Doubles; and
org.apache.spark.rdd.SequenceFileRDDFunctions contains operations available on RDDs that can
be saved as SequenceFiles. These operations are automatically available on any RDD of the right
type (e.g. RDD[(Int, Int)] through implicit conversions.
Java programmers should reference the org.apache.spark.api.java package for Spark programming APIs in Java.
Classes and methods marked with Experimental are user-facing features which have not been officially adopted by the Spark project. These are subject to change or removal in minor releases.
Classes and methods marked with Developer API are intended for advanced users want to extend Spark through lower level interfaces. These are subject to changes or removal in minor releases.
DataFrame-based machine learning APIs to let users quickly assemble and configure practical machine learning pipelines.
DataFrame-based machine learning APIs to let users quickly assemble and configure practical machine learning pipelines.
The ML pipeline API uses DataFrames as ML datasets.
The ML pipeline API uses DataFrames as ML datasets.
Each dataset consists of typed columns, e.g., string, double, vector, etc.
However, knowing only the column type may not be sufficient to handle the data properly.
For instance, a double column with values 0.0, 1.0, 2.0, ... may represent some label indices,
which cannot be treated as numeric values in ML algorithms, and, for another instance, we may
want to know the names and types of features stored in a vector column.
ML attributes are used to provide additional information to describe columns in a dataset.
A column with ML attributes attached is called an ML column. The data in ML columns are stored as double values, i.e., an ML column is either a scalar column of double values or a vector column. Columns of other types must be encoded into ML columns using transformers. We use Attribute to describe a scalar ML column, and AttributeGroup to describe a vector ML column. ML attributes are stored in the metadata field of the column schema.
The ml.feature package provides common feature transformers that help convert raw data or
features into more suitable forms for model fitting.
The ml.feature package provides common feature transformers that help convert raw data or
features into more suitable forms for model fitting.
Most feature transformers are implemented as Transformers, which transform one DataFrame
into another, e.g., HashingTF.
Some feature transformers are implemented as Estimators, because the transformation requires
some aggregated information of the dataset, e.g., document frequencies in IDF.
For those feature transformers, calling Estimator.fit is required to obtain the model first,
e.g., IDFModel, in order to apply transformation.
The transformation is usually done by appending new columns to the input DataFrame, so all
input columns are carried over.
We try to make each transformer minimal, so it becomes flexible to assemble feature transformation pipelines. Pipeline can be used to chain feature transformers, and VectorAssembler can be used to combine multiple feature transformations, for example:
import org.apache.spark.ml.feature._ import org.apache.spark.ml.Pipeline // a DataFrame with three columns: id (integer), text (string), and rating (double). val df = spark.createDataFrame(Seq( (0, "Hi I heard about Spark", 3.0), (1, "I wish Java could use case classes", 4.0), (2, "Logistic regression models are neat", 4.0) )).toDF("id", "text", "rating") // define feature transformers val tok = new RegexTokenizer() .setInputCol("text") .setOutputCol("words") val sw = new StopWordsRemover() .setInputCol("words") .setOutputCol("filtered_words") val tf = new HashingTF() .setInputCol("filtered_words") .setOutputCol("tf") .setNumFeatures(10000) val idf = new IDF() .setInputCol("tf") .setOutputCol("tf_idf") val assembler = new VectorAssembler() .setInputCols(Array("tf_idf", "rating")) .setOutputCol("features") // assemble and fit the feature transformation pipeline val pipeline = new Pipeline() .setStages(Array(tok, sw, tf, idf, assembler)) val model = pipeline.fit(df) // save transformed features with raw data model.transform(df) .select("id", "text", "rating", "features") .write.format("parquet").save("/output/path")
Some feature transformers implemented in MLlib are inspired by those implemented in scikit-learn. The major difference is that most scikit-learn feature transformers operate eagerly on the entire input dataset, while MLlib's feature transformers operate lazily on individual columns, which is more efficient and flexible to handle large and complex datasets.