I am still using Spark 1.6 and trained a StandardScalar that I would like to save and reuse on future datasets.
Using the supplied examples I could transform the data successfully but I can't find a way to save the trained normaliser.
Is there any way in which the trained normaliser can be saved?
Assuming that you have created the scalerModel:
import org.apache.spark.ml.feature.StandardScalerModel
scalerModel.write.save("path/folder/")
val scalerModel = StandardScalerModel.load("path/folder/")
StandardScalerModel class has a save method. After calling the fit method on StandardScaler, the returned object is StandardScalerModel: API Docs
e.g. similar to the supplied example:
import org.apache.spark.ml.feature.StandardScaler
import org.apache.spark.ml.PipelineModel
val dataFrame = spark.read.format("libsvm").load("data/mllib/sample_libsvm_data.txt")
val scaler = new StandardScaler()
.setInputCol("features")
.setOutputCol("scaledFeatures")
.setWithStd(true)
.setWithMean(false)
// Compute summary statistics by fitting the StandardScaler.
val scalerModel = scaler.fit(dataFrame)
scalerModel.write.overwrite().save("/path/to/the/file")
val sameModel = PipelineModel.load("/path/to/the/file")
Related
I am writing an Apache Spark application using Scala. To handle and store data I use DataFrames. I have a nice pipeline with feature extraction and a MultiLayerPerceptron classifier, using the ML API.
I also want to use SVM (for comparison purposes). The thing is (and correct me if I am mistaken) only the MLLib provides SVM. And MLLib is not ready to handle DataFrames, only RDDs.
So I figured I can maintain the core of my application using DataFrames and to use SVM 1) I just convert the DataFrame's columns I need to an RDD[LabeledPoint] and 2) after the classification add the SVMs prediction to the DataFrame as a new column.
The first part I handled with a small function:
private def dataFrameToRDD(dataFrame : DataFrame) : RDD[LabeledPoint] = {
val rddMl = dataFrame.select("label", "features").rdd.map(r => (r.getInt(0).toDouble, r.getAs[org.apache.spark.ml.linalg.SparseVector](1)))
rddMl.map(r => new LabeledPoint(r._1, Vectors.dense(r._2.toArray)))
}
I have to specify and convert the type of vector since the feature extraction method uses ML API and not MLLib.
Then, this RDD[LabeledPoint] is fed to the SVM and classification goes smoothly, no issues. At the end and following spark's example I get an RDD[Double]:
val predictions = rdd.map(point => model.predict(point.features))
Now, I want to add the prediction score as column to the original DataFrame and return it. This is where I got stuck. I can convert the RDD[Double] to a DataFrame using
(sql context ommited)
import sqlContext.implicits._
val plDF = predictions.toDF("prediction")
But how do I join the two DataFrames where the second DataFrame becomes a column of the original one? I tried to use methods join and union but got SQL exceptions as the DataFrames have no equal columns to join or unite on.
EDIT
I tried
data.withColumn("prediction", plDF.col("prediction"))
But I get an Analysis Exception :(
I haven't figured out how to do it without recurring to RDDs, but anyway here's how I solved it with RDD. Added the rest of the code so that anyone can understand the complete logic. Any suggestions are appreciated.
package stuff
import java.util.logging.{Level, Logger}
import org.apache.spark.mllib.classification.{SVMModel, SVMWithSGD}
import org.apache.spark.mllib.linalg.Vectors
import org.apache.spark.mllib.regression.LabeledPoint
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.{DataFrame, Row, SQLContext}
/**
* Created by camandros on 10-03-2017.
*/
class LinearSVMClassifier extends Classifier with Serializable{
#transient lazy val log: Logger = Logger.getLogger(getClass.getName)
private var model : SVMModel = _
override def train(data : DataFrame): Unit = {
val rdd = dataFrameToRDD(data)
// Run training algorithm to build the model
val numIter : Int = 100
val step = Osint.properties(Osint.SVM_STEPSIZE).toDouble
val c = Osint.properties(Osint.SVM_C).toDouble
log.log(Level.INFO, "Initiating SVM training with parameters: C="+c+", step="+step)
model = SVMWithSGD.train(rdd, numIterations = numIter, stepSize = step, regParam = c)
log.log(Level.INFO, "Model training finished")
// Clear the default threshold.
model.clearThreshold()
}
override def classify(data : DataFrame): DataFrame = {
log.log(Level.INFO, "Converting DataFrame to RDD")
val rdd = dataFrameToRDD(data)
log.log(Level.INFO, "Conversion finished; beginning classification")
// Compute raw scores on the test set.
val predictions = rdd.map(point => model.predict(point.features))
log.log(Level.INFO, "Classification finished; Transforming RDD to DataFrame")
val sqlContext : SQLContext = Osint.spark.sqlContext
val tupleRDD = data.rdd.zip(predictions).map(t => Row.fromSeq(t._1.toSeq ++ Seq(t._2)))
sqlContext.createDataFrame(tupleRDD, data.schema.add("predictions", "Double"))
//TODO this should work it doesn't since this "withColumn" method seems to be applicable only to add
// new columns using information from the same dataframe; therefore I am using the horrible rdd conversion
//val sqlContext : SQLContext = Osint.spark.sqlContext
//import sqlContext.implicits._
//val plDF = predictions.toDF("predictions")
//data.withColumn("prediction", plDF.col("predictions"))
}
private def dataFrameToRDD(dataFrame : DataFrame) : RDD[LabeledPoint] = {
val rddMl = dataFrame.select("label", "features").rdd.map(r => (r.getInt(0).toDouble, r.getAs[org.apache.spark.ml.linalg.SparseVector](1)))
rddMl.map(r => new LabeledPoint(r._1, Vectors.dense(r._2.toArray)))
}
}
I have an RDD[LabeledPoint] intended to be used within a machine learning pipeline. How do we convert that RDD to a DataSet? Note the newer spark.ml apis require inputs in the Dataset format.
Here is an answer that traverses an extra step - the DataFrame. We use the SQLContext to create a DataFrame and then create a DataSet using the desired object type - in this case a LabeledPoint:
val sqlContext = new SQLContext(sc)
val pointsTrainDf = sqlContext.createDataFrame(training)
val pointsTrainDs = pointsTrainDf.as[LabeledPoint]
Update Ever heard of a SparkSession ? (neither had I until now..)
So apparently the SparkSession is the Preferred Way (TM) in Spark 2.0.0 and moving forward. Here is the updated code for the new (spark) world order:
Spark 2.0.0+ approaches
Notice in both of the below approaches (simpler one of which credit #zero323) we have accomplished an important savings as compared to the SQLContext approach: no longer is it necessary to first create a DataFrame.
val sparkSession = SparkSession.builder().getOrCreate()
val pointsTrainDf = sparkSession.createDataset(training)
val model = new LogisticRegression()
.train(pointsTrainDs.as[LabeledPoint])
Second way for Spark 2.0.0+ Credit to #zero323
val spark: org.apache.spark.sql.SparkSession = ???
import spark.implicits._
val trainDs = training.toDS()
Traditional Spark 1.X and earlier approach
val sqlContext = new SQLContext(sc) // Note this is *deprecated* in 2.0.0
import sqlContext.implicits._
val training = splits(0).cache()
val test = splits(1)
val trainDs = training**.toDS()**
See also: How to store custom objects in Dataset? by the esteemed #zero323 .
I built a random forest model using the following code:
import org.apache.spark.ml.classification.RandomForestClassificationModel
import org.apache.spark.ml.classification.RandomForestClassifier
val rf = new RandomForestClassifier().setLabelCol("indexedLabel").setFeaturesCol("features")
val labelConverter = new IndexToString().setInputCol("prediction").setOutputCol("predictedLabel").setLabels(labelIndexer.labels)
val training = labelIndexer.transform(df)
val model = rf.fit(training)
now I want to save the model in order to predict later using the following code:
val predictions: DataFrame = model.transform(testData)
I've looked into Spark documentation here and didn't find any option to do that. Any idea?
It took me a few hours to build the model , if Spark is crushing I won't be able to get it back.
It's possible to save and reload tree based models in HDFS using Spark 1.6 using saveAsObjectFile() for both Pipeline based and basic model.
Below is example for pipeline based model.
// model
val model = pipeline.fit(trainingData)
// Create rdd using Seq
sc.parallelize(Seq(model), 1).saveAsObjectFile("hdfs://filepath")
// Reload model by using it's class
// You can get class of object using object.getClass()
val sameModel = sc.objectFile[PipelineModel]("filepath").first()
For RandomForestClassifier save & load model: tested spark 1.6.2 + scala in ml(in spark 2.0 you can have direct save option for model)
import org.apache.spark.ml.classification.RandomForestClassificationModel
import org.apache.spark.ml.classification.RandomForestClassifier //imports
val classifier = new RandomForestClassifier().setImpurity("gini").setMaxDepth(3).setNumTrees(20).setFeatureSubsetStrategy("auto").setSeed(5043)
val model = classifier.fit(trainingData)
sc.parallelize(Seq(model), 1).saveAsObjectFile(modelSavePath) //save model
val linRegModel = sc.objectFile[RandomForestClassificationModel](modelSavePath).first() //load model
`val predictions1 = linRegModel.transform(testData)` //predictions1 is dataframe
It is in the MLWriter interface - that is accessed via the writer attribute on your model:
model.asInstanceOf[MLWritable].write.save(path)
Here is the interface:
abstract class MLWriter extends BaseReadWrite with Logging {
protected var shouldOverwrite: Boolean = false
/**
* Saves the ML instances to the input path.
*/
#Since("1.6.0")
#throws[IOException]("If the input path already exists but overwrite is not enabled.")
def save(path: String): Unit = {
This is a refactoring from earlier versions of mllib/spark.ml
Update It appears that the Model were not writable:
Exception in thread "main" java.lang.UnsupportedOperationException:
Pipeline write will fail on this Pipeline because it contains a stage
which does not implement Writable. Non-Writable stage:
rfc_4e467607406f of type class
org.apache.spark.ml.classification.RandomForestClassificationModel
So there may not be a straightforward solution for this.
Here is a PySpark v1.6 implementation corresponding to the Scala saveAsObjectFile() answer above.
It coerses the Python objects to/from Java objects to achieve serialisation with saveAsObjectFile().
Without the Java coersion I had weird Py4J errors on serialisation. If anyone has a simplier implementation, please edit or comment.
Save a trained RandomForestClassificationModel object:
# Save RandomForestClassificationModel to hdfs
gateway = sc._gateway
java_list = gateway.jvm.java.util.ArrayList()
java_list.add(rfModel._java_obj)
modelRdd = sc._jsc.parallelize(java_list)
modelRdd.saveAsObjectFile("hdfs:///some/path/rfModel")
Load a trained RandomForestClassificationModel object:
# Load RandomForestClassificationModel from hdfs
rfObjectFileLoaded = sc._jsc.objectFile("hdfs:///some/path/rfModel")
rfModelLoaded_JavaObject = rfObjectFileLoaded.first()
rfModelLoaded = RandomForestClassificationModel(rfModelLoaded_JavaObject)
predictions = rfModelLoaded.transform(test_input_df)
Hi I am new to spark and scala. I am running scala code in spark scala prompt. The program is fine, it's showing "defined module MLlib" but its not printing anything on screen. What have I done wrong? Is there any other way to run this program spark in scala shell and get the output?
import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.mllib.classification.LogisticRegressionWithSGD
import org.apache.spark.mllib.feature.HashingTF
import org.apache.spark.mllib.regression.LabeledPoint
object MLlib {
def main(args: Array[String]) {
val conf = new SparkConf().setAppName(s"Book example: Scala")
val sc = new SparkContext(conf)
// Load 2 types of emails from text files: spam and ham (non-spam).
// Each line has text from one email.
val spam = sc.textFile("/home/training/Spam.txt")
val ham = sc.textFile("/home/training/Ham.txt")
// Create a HashingTF instance to map email text to vectors of 100 features.
val tf = new HashingTF(numFeatures = 100)
// Each email is split into words, and each word is mapped to one feature.
val spamFeatures = spam.map(email => tf.transform(email.split(" ")))
val hamFeatures = ham.map(email => tf.transform(email.split(" ")))
// Create LabeledPoint datasets for positive (spam) and negative (ham) examples.
val positiveExamples = spamFeatures.map(features => LabeledPoint(1, features))
val negativeExamples = hamFeatures.map(features => LabeledPoint(0, features))
val trainingData = positiveExamples ++ negativeExamples
trainingData.cache() // Cache data since Logistic Regression is an iterative algorithm.
// Create a Logistic Regression learner which uses the LBFGS optimizer.
val lrLearner = new LogisticRegressionWithSGD()
// Run the actual learning algorithm on the training data.
val model = lrLearner.run(trainingData)
// Test on a positive example (spam) and a negative one (ham).
// First apply the same HashingTF feature transformation used on the training data.
val posTestExample = tf.transform("O M G GET cheap stuff by sending money to ...".split(" "))
val negTestExample = tf.transform("Hi Dad, I started studying Spark the other ...".split(" "))
// Now use the learned model to predict spam/ham for new emails.
println(s"Prediction for positive test example: ${model.predict(posTestExample)}")
println(s"Prediction for negative test example: ${model.predict(negTestExample)}")
sc.stop()
}
}
A couple of things:
You defined your object in the the Spark shell, so the main class won't get called immediately. You'll have to call it explicitly after you define the object:
MLlib.main(Array())
In fact, if you continue to work on the shell/REPL you can do away with the object altogether; you can define the function directly. For example:
import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.mllib.classification.LogisticRegressionWithSGD
import org.apache.spark.mllib.feature.HashingTF
import org.apache.spark.mllib.regression.LabeledPoint
def MLlib {
//the rest of your code
}
However, you shouldn't initialize SparkContext it within the shell. From the documentation:
In the Spark shell, a special interpreter-aware SparkContext is
already created for you, in the variable called sc. Making your own
SparkContext will not work
So, you have to either remove that bit from your code, or compile it into a jar and run it using spark-submit
I've been able to create a pipeline that will allow me to index multiple string columns at once, but I am getting stuck encoding them, because unlike indexing, the encoder is not an estimator so I never call fit
according to the OneHotEncoder example in the docs.
import org.apache.spark.ml.feature.{StringIndexer, VectorAssembler,
OneHotEncoder}
import org.apache.spark.ml.Pipeline
val data = sqlContext.read.parquet("s3n://map2-test/forecaster/intermediate_data")
val df = data.select("win","bid_price","domain","size", "form_factor").na.drop()
//indexing columns
val stringColumns = Array("domain","size", "form_factor")
val index_transformers: Array[org.apache.spark.ml.PipelineStage] = stringColumns.map(
cname => new StringIndexer()
.setInputCol(cname)
.setOutputCol(s"${cname}_index")
)
// Add the rest of your pipeline like VectorAssembler and algorithm
val index_pipeline = new Pipeline().setStages(index_transformers)
val index_model = index_pipeline.fit(df)
val df_indexed = index_model.transform(df)
//encoding columns
val indexColumns = df_indexed.columns.filter(x => x contains "index")
val one_hot_encoders: Array[org.apache.spark.ml.PipelineStage] = indexColumns.map(
cname => new OneHotEncoder()
.setInputCol(cname)
.setOutputCol(s"${cname}_vec")
)
val one_hot_pipeline = new Pipeline().setStages(one_hot_encoders)
val df_encoded = one_hot_pipeline.transform(df_indexed)
The OneHotEncoder object doesn't have a fit method so putting it in the same pipeline as the indexers will not work- it throws an error when I call fit on the pipeline. I can also not call transform on the pipeline that I made with the array of pipeline stages, one_hot_encoders.
I have not found a good solution for using the OneHotEncoder without individually creating and calling transform on that transforming itself for all of the columns I want to encode
Spark >= 3.0:
In Spark 3.0 OneHotEncoderEstimator has been renamed to OneHotEncoder:
import org.apache.spark.ml.feature.{OneHotEncoder, OneHotEncoderModel}
val encoder = new OneHotEncoder()
.setInputCols(indexColumns)
.setOutputCols(indexColumns map (name => s"${name}_vec"))
Spark >= 2.3
Spark 2.3 introduced new classes OneHotEncoderEstimator, OneHotEncoderModel, which required fitting even if used outside Pipeline, and operate on multiple columns at the same time.
import org.apache.spark.ml.feature.{OneHotEncoderEstimator, OneHotEncoderModel}
val encoder = new OneHotEncoderEstimator()
.setInputCols(indexColumns)
.setOutputCols(indexColumns map (name => s"${name}_vec"))
encoder.fit(df_indexed).transform(df_indexed)
Spark < 2.3
Even if transformers you use don't require fitting you have to use fit method to create a PipelineModel which can be used to transform data.
one_hot_pipeline.fit(df_indexed).transform(df_indexed)
On a side note you can combine indexing and encoding into a single Pipeline:
val pipeline = new Pipeline()
.setStages(index_transformers ++ one_hot_encoders)
val model = pipeline.fit(df)
model.transform(df)
Edit:
Error you see means that one of your columns contains an empty String. It is accepted by indexer but cannot be used for encoding. Depending on you requirements you can drop these or use a dummy label. Unfortunately you cannot use NULLs until SPARK-11569) is resolved.