modules/ml/src/main/java/org/apache/ignite/ml/preprocessing/minmaxscaling/MinMaxScalerPreprocessor.java

   1 /*
   2  * Licensed to the Apache Software Foundation (ASF) under one or more
   3  * contributor license agreements.  See the NOTICE file distributed with
   4  * this work for additional information regarding copyright ownership.
   5  * The ASF licenses this file to You under the Apache License, Version 2.0
   6  * (the "License"); you may not use this file except in compliance with
   7  * the License.  You may obtain a copy of the License at
   8  *
   9  *      http://www.apache.org/licenses/LICENSE-2.0
  10  *
  11  * Unless required by applicable law or agreed to in writing, software
  12  * distributed under the License is distributed on an "AS IS" BASIS,
  13  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  14  * See the License for the specific language governing permissions and
  15  * limitations under the License.
  16  */
  17
  18 package org.apache.ignite.ml.preprocessing.minmaxscaling;
  19
  20 import org.apache.ignite.ml.math.Vector;
  21 import org.apache.ignite.ml.math.functions.IgniteBiFunction;
  22
  23 /**
  24  * Preprocessing function that makes minmaxscaling. From mathematical point of view it's the following function which
  25  * is applied to every element in dataset:
  26  *
  27  * {@code a_i = (a_i - min_i) / (max_i - min_i) for all i},
  28  *
  29  * where {@code i} is a number of column, {@code max_i} is the value of the maximum element in this columns,
  30  * {@code min_i} is the value of the minimal element in this column.
  31  *
  32  * @param <K> Type of a key in {@code upstream} data.
  33  * @param <V> Type of a value in {@code upstream} data.
  34  */
  35 public class MinMaxScalerPreprocessor<K, V> implements IgniteBiFunction<K, V, Vector> {
  36     /** */
  37     private static final long serialVersionUID = 6997800576392623469L;
  38
  39     /** Minimal values. */
  40     private final double[] min;
  41
  42     /** Maximum values. */
  43     private final double[] max;
  44
  45     /** Base preprocessor. */
  46     private final IgniteBiFunction<K, V, Vector> basePreprocessor;
  47
  48     /**
  49      * Constructs a new instance of minmaxscaling preprocessor.
  50      *
  51      * @param min Minimal values.
  52      * @param max Maximum values.
  53      * @param basePreprocessor Base preprocessor.
  54      */
  55     public MinMaxScalerPreprocessor(double[] min, double[] max, IgniteBiFunction<K, V, Vector> basePreprocessor) {
  56         this.min = min;
  57         this.max = max;
  58         this.basePreprocessor = basePreprocessor;
  59     }
  60
  61     /**
  62      * Applies this preprocessor.
  63      *
  64      * @param k Key.
  65      * @param v Value.
  66      * @return Preprocessed row.
  67      */
  68     @Override public Vector apply(K k, V v) {
  69         Vector res = basePreprocessor.apply(k, v);
  70
  71         assert res.size() == min.length;
  72         assert res.size() == max.length;
  73
  74         for (int i = 0; i < res.size(); i++)
  75             res.set(i, (res.get(i) - min[i]) / (max[i] - min[i]));
  76
  77         return res;
  78     }
  79
  80     /** */
  81     public double[] getMin() {
  82         return min;
  83     }
  84
  85     /** */
  86     public double[] getMax() {
  87         return max;
  88     }
  89 }