Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Feature normalization algorithm in Spark

Trying to understand Spark's normalization algorithm. My small test set contains 5 vectors:

{0.95, 0.018, 0.0, 24.0, 24.0, 14.4, 70000.0},  
{1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 70000.0},  
{-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, 70000.0},  
{-0.95, 0.018, 0.0, 24.0, 24.0, 14.4, 70000.0},  
{0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 70000.0},  

I would expect that new Normalizer().transform(vectors) creates JavaRDD where each vector feature is normalized as (v-mean)/stdev across all values for feature-0, `feature-1, etc.
The resulting set is:

[-1.4285714276967932E-5,-1.4285714276967932E-5,-1.4285714276967932E-5,-1.4285714276967932E-5,-1.4285714276967932E-5,-1.4285714276967932E-5,0.9999999993877552]  
[1.357142668768307E-5,2.571428214508371E-7,0.0,3.428570952677828E-4,3.428570952677828E-4,2.057142571606697E-4,0.9999998611976999]  
[-1.357142668768307E-5,2.571428214508371E-7,0.0,3.428570952677828E-4,3.428570952677828E-4,2.057142571606697E-4,0.9999998611976999]  
[1.4285714276967932E-5,1.4285714276967932E-5,1.4285714276967932E-5,1.4285714276967932E-5,1.4285714276967932E-5,1.4285714276967932E-5,0.9999999993877552]  
[0.0,0.0,0.0,0.0,0.0,0.0,1.0]  

Note that all original values 7000.0 result in different 'normalized' values. Also, how, for example, 1.357142668768307E-5 was calculated when the values are: .95, 1,-1, -.95, 0? What's more, if I remove a feature, the results are different. Could not find any documentation on the issue.
In fact, my question is, how to normalize all vectors in RDD correctly?

like image 363
Alex B Avatar asked Dec 12 '15 00:12

Alex B


People also ask

What is normalization in Spark?

Normalizer - Data Science with Apache Spark. Normalizer. Normalizer is a Transformer which transforms a dataset of Vector rows, normalizing each Vector to have unit norm.

Why do we use VectorAssembler in PySpark?

VectorAssembler is a transformer that combines a given list of columns into a single vector column. It is useful for combining raw features and features generated by different feature transformers into a single feature vector, in order to train ML models like logistic regression and decision trees.

What is feature engineering in Spark?

March 30, 2021. Apache Spark MLlib contains many utility functions for performing feature engineering at scale, including methods for encoding and transforming features. These methods can also be used to process features for other machine learning libraries.

How do you do feature selection in PySpark?

Currently, Spark supports three Univariate Feature Selectors: chi-squared, ANOVA F-test and F-value. User can choose Univariate Feature Selector by setting featureType and labelType , and Spark will pick the score function based on the specified featureType and labelType .


1 Answers

Your expectations are simply incorrect. As it is clearly stated in the official documentation "Normalizer scales individual samples to have unit L p norm" where default value for p is 2. Ignoring numerical precision issues:

import org.apache.spark.mllib.linalg.Vectors

val rdd = sc.parallelize(Seq(
    Vectors.dense(0.95, 0.018, 0.0, 24.0, 24.0, 14.4, 70000.0),  
    Vectors.dense(1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 70000.0),  
    Vectors.dense(-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, 70000.0),  
    Vectors.dense(-0.95, 0.018, 0.0, 24.0, 24.0, 14.4, 70000.0),  
    Vectors.dense(0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 70000.0)))

val transformed = normalizer.transform(rdd)
transformed.map(_.toArray.sum).collect
// Array[Double] = Array(1.0009051182149054, 1.000085713673417,
//   0.9999142851020933, 1.00087797536153, 1.0

MLLib doesn't provide functionality you need but can use StandardScaler from ML.

import org.apache.spark.ml.feature.StandardScaler

val df = rdd.map(Tuple1(_)).toDF("features")

val scaler = new StandardScaler()
  .setInputCol("features")
  .setOutputCol("scaledFeatures")
  .setWithStd(true)
  .setWithMean(true)

val transformedDF =  scaler.fit(df).transform(df)

transformedDF.select($"scaledFeatures")show(5, false)

// +--------------------------------------------------------------------------------------------------------------------------+
// |scaledFeatures                                                                                                            |
// +--------------------------------------------------------------------------------------------------------------------------+
// |[0.9740388301169303,0.015272022105217588,0.0,1.0938637007095298,1.0938637007095298,1.0910691283447955,0.0]                |
// |[1.0253040317020319,1.4038947727833362,1.414213562373095,-0.6532797101459693,-0.6532797101459693,-0.6010982697825494,0.0] |
// |[-1.0253040317020319,-1.4242574689236265,-1.414213562373095,-0.805205224133404,-0.805205224133404,-0.8536605680105113,0.0]|
// |[-0.9740388301169303,0.015272022105217588,0.0,1.0938637007095298,1.0938637007095298,1.0910691283447955,0.0]               |
// |[0.0,-0.010181348070145075,0.0,-0.7292424671396867,-0.7292424671396867,-0.7273794188965303,0.0]                           |
// +--------------------------------------------------------------------------------------------------------------------------+
like image 56
zero323 Avatar answered Oct 14 '22 01:10

zero323