Start line:  
End line:  

Snippet Preview

Snippet HTML Code

Stack Overflow Questions
  /*
   * Copyright 2009-2013 by The Regents of the University of California
   * Licensed under the Apache License, Version 2.0 (the "License");
   * you may not use this file except in compliance with the License.
   * you may obtain a copy of the License from
   * 
   *     http://www.apache.org/licenses/LICENSE-2.0
   * 
   * Unless required by applicable law or agreed to in writing, software
  * distributed under the License is distributed on an "AS IS" BASIS,
  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  * See the License for the specific language governing permissions and
  * limitations under the License.
  */
Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership. The ASF licenses this file to you under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
 
 package org.apache.hadoop.hive.ql.udf.generic;
 
 
 import  org.apache.hadoop.hive.ql.exec.Description;
 import  org.apache.hadoop.hive.ql.exec.UDFArgumentTypeException;
 import  org.apache.hadoop.hive.ql.metadata.HiveException;
 import  org.apache.hadoop.hive.ql.parse.SemanticException;
 import  org.apache.hadoop.hive.serde2.io.DoubleWritable;
 import  org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector;
 import  org.apache.hadoop.hive.serde2.objectinspector.ObjectInspectorFactory;
 import  org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector;
 import  org.apache.hadoop.hive.serde2.objectinspector.StructField;
 import  org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector;
 import  org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector;
 import  org.apache.hadoop.hive.serde2.objectinspector.primitive.LongObjectInspector;
 import  org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorFactory;
 import  org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorUtils;
 import  org.apache.hadoop.hive.serde2.typeinfo.PrimitiveTypeInfo;
 import  org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
 import  org.apache.hadoop.io.LongWritable;
 
Compute the covariance covar_pop(x, y), using the following one-pass method (ref. "Formulas for Robust, One-Pass Parallel Computation of Covariances and Arbitrary-Order Statistical Moments", Philippe Pebay, Sandia Labs): Incremental: n : <count> mx_n = mx_(n-1) + [x_n - mx_(n-1)]/n : <xavg> my_n = my_(n-1) + [y_n - my_(n-1)]/n : <yavg> c_n = c_(n-1) + (x_n - mx_(n-1))*(y_n - my_n) : <covariance * n> Merge: c_X = c_A + c_B + (mx_A - mx_B)*(my_A - my_B)*n_A*n_B/n_X
 
 @Description(name = "covariance,covar_pop", value = "_FUNC_(x,y) - Returns the population covariance of a set of number pairs", extended = "The function takes as arguments any pair of numeric types and returns a double.\n"
         + "Any pair with a NULL is ignored. If the function is applied to an empty set, NULL\n"
         + "will be returned. Otherwise, it computes the following:\n"
         + "   (SUM(x*y)-SUM(x)*SUM(y)/COUNT(x,y))/COUNT(x,y)\n" + "where neither x nor y is null.")
 public class GenericUDAFCovariance extends AbstractGenericUDAFResolver {
 
     static final Log LOG = LogFactory.getLog(GenericUDAFCovariance.class.getName());
 
     @Override
     public GenericUDAFEvaluator getEvaluator(TypeInfo[] parametersthrows SemanticException {
         if (parameters.length != 2) {
             throw new UDFArgumentTypeException(parameters.length - 1, "Exactly two arguments are expected.");
         }
 
         if (parameters[0].getCategory() != ObjectInspector.Category.PRIMITIVE) {
             throw new UDFArgumentTypeException(0, "Only primitive type arguments are accepted but "
                     + parameters[0].getTypeName() + " is passed.");
         }
 
         if (parameters[1].getCategory() != ObjectInspector.Category.PRIMITIVE) {
             throw new UDFArgumentTypeException(1, "Only primitive type arguments are accepted but "
                     + parameters[1].getTypeName() + " is passed.");
         }
 
         switch (((PrimitiveTypeInfo) parameters[0]).getPrimitiveCategory()) {
             case BYTE:
             case SHORT:
             case INT:
             case LONG:
             case FLOAT:
            case DOUBLE:
                switch (((PrimitiveTypeInfo) parameters[1]).getPrimitiveCategory()) {
                    case BYTE:
                    case SHORT:
                    case INT:
                    case LONG:
                    case FLOAT:
                    case DOUBLE:
                        return new GenericUDAFCovarianceEvaluator();
                    case STRING:
                    case BOOLEAN:
                    default:
                        throw new UDFArgumentTypeException(1, "Only numeric or string type arguments are accepted but "
                                + parameters[1].getTypeName() + " is passed.");
                }
            case STRING:
            case BOOLEAN:
            default:
                throw new UDFArgumentTypeException(0, "Only numeric or string type arguments are accepted but "
                        + parameters[0].getTypeName() + " is passed.");
        }
    }

    
Evaluate the variance using the algorithm described in http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance, presumably by P├ębay, Philippe (2008), in "Formulas for Robust, One-Pass Parallel Computation of Covariances and Arbitrary-Order Statistical Moments", Technical Report SAND2008-6212, Sandia National Laboratories, http://infoserve.sandia.gov/sand_doc/2008/086212.pdf Incremental: n : <count> mx_n = mx_(n-1) + [x_n - mx_(n-1)]/n : <xavg> my_n = my_(n-1) + [y_n - my_(n-1)]/n : <yavg> c_n = c_(n-1) + (x_n - mx_(n-1))*(y_n - my_n) : <covariance * n> Merge: c_X = c_A + c_B + (mx_A - mx_B)*(my_A - my_B)*n_A*n_B/n_X This one-pass algorithm is stable.
    public static class GenericUDAFCovarianceEvaluator extends GenericUDAFEvaluator {
        // For PARTIAL1 and COMPLETE
        private PrimitiveObjectInspector xInputOI;
        private PrimitiveObjectInspector yInputOI;
        // For PARTIAL2 and FINAL
        private StructObjectInspector soi;
        private StructField countField;
        private StructField xavgField;
        private StructField yavgField;
        private StructField covarField;
        private LongObjectInspector countFieldOI;
        private DoubleObjectInspector xavgFieldOI;
        private DoubleObjectInspector yavgFieldOI;
        private DoubleObjectInspector covarFieldOI;
        // For PARTIAL1 and PARTIAL2
        private Object[] partialResult;
        // For FINAL and COMPLETE
        private DoubleWritable result;
        @Override
        public ObjectInspector init(Mode m, ObjectInspector[] parametersthrows HiveException {
            super.init(mparameters);
            // init input
            if (mode == Mode.PARTIAL1 || mode == Mode.COMPLETE) {
                assert (parameters.length == 2);
                 = (PrimitiveObjectInspector) parameters[0];
                 = (PrimitiveObjectInspector) parameters[1];
            } else {
                assert (parameters.length == 1);
                 = (StructObjectInspector) parameters[0];
                 = .getStructFieldRef("count");
                 = .getStructFieldRef("xavg");
                 = .getStructFieldRef("yavg");
                 = .getStructFieldRef("covar");
                 = (LongObjectInspector) .getFieldObjectInspector();
                 = (DoubleObjectInspector) .getFieldObjectInspector();
                 = (DoubleObjectInspector) .getFieldObjectInspector();
                 = (DoubleObjectInspector) .getFieldObjectInspector();
            }
            // init output
            if (mode == Mode.PARTIAL1 || mode == Mode.PARTIAL2) {
                // The output of a partial aggregation is a struct containing
                // a long count, two double averages, and a double covariance.
                ArrayList<ObjectInspector> foi = new ArrayList<ObjectInspector>();
                foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
                foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
                foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
                foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
                ArrayList<Stringfname = new ArrayList<String>();
                fname.add("count");
                fname.add("xavg");
                fname.add("yavg");
                fname.add("covar");
                 = new Object[4];
                [0] = new LongWritable(0);
                [1] = new DoubleWritable(0);
                [2] = new DoubleWritable(0);
                [3] = new DoubleWritable(0);
                return ObjectInspectorFactory.getStandardStructObjectInspector(fnamefoi);
            } else {
                setResult(new DoubleWritable(0));
                return PrimitiveObjectInspectorFactory.writableDoubleObjectInspector;
            }
        }
        static class StdAgg implements SerializableBuffer {
            long count// number n of elements
            double xavg// average of x elements
            double yavg// average of y elements
            double covar// n times the covariance
            @Override
            public void deSerializeAggBuffer(byte[] dataint startint len) {
                 = BufferSerDeUtil.getLong(datastart);
                start += 8;
                 = BufferSerDeUtil.getDouble(datastart);
                start += 8;
                 = BufferSerDeUtil.getDouble(datastart);
                start += 8;
                 = BufferSerDeUtil.getDouble(datastart);
            }
            @Override
            public void serializeAggBuffer(byte[] dataint startint len) {
                BufferSerDeUtil.writeLong(datastart);
                start += 8;
                BufferSerDeUtil.writeDouble(datastart);
                start += 8;
                BufferSerDeUtil.writeDouble(datastart);
                start += 8;
                BufferSerDeUtil.writeDouble(datastart);
            }
            @Override
            public void serializeAggBuffer(DataOutput outputthrows IOException {
                output.writeLong();
                output.writeDouble();
                output.writeDouble();
                output.writeDouble();
            }
        };
        @Override
        public AggregationBuffer getNewAggregationBuffer() throws HiveException {
            StdAgg result = new StdAgg();
            reset(result);
            return result;
        }
        @Override
        public void reset(AggregationBuffer aggthrows HiveException {
            StdAgg myagg = (StdAggagg;
            myagg.count = 0;
            myagg.xavg = 0;
            myagg.yavg = 0;
            myagg.covar = 0;
        }
        @Override
        public void iterate(AggregationBuffer aggObject[] parametersthrows HiveException {
            assert (parameters.length == 2);
            Object px = parameters[0];
            Object py = parameters[1];
            if (px != null && py != null) {
                StdAgg myagg = (StdAggagg;
                double vx = PrimitiveObjectInspectorUtils.getDouble(px);
                double vy = PrimitiveObjectInspectorUtils.getDouble(py);
                myagg.count++;
                myagg.yavg = myagg.yavg + (vy - myagg.yavg) / myagg.count;
                if (myagg.count > 1) {
                    myagg.covar += (vx - myagg.xavg) * (vy - myagg.yavg);
                }
                myagg.xavg = myagg.xavg + (vx - myagg.xavg) / myagg.count;
            }
        }
        @Override
        public Object terminatePartial(AggregationBuffer aggthrows HiveException {
            StdAgg myagg = (StdAggagg;
            ((LongWritable) [0]).set(myagg.count);
            ((DoubleWritable) [1]).set(myagg.xavg);
            ((DoubleWritable) [2]).set(myagg.yavg);
            ((DoubleWritable) [3]).set(myagg.covar);
            return ;
        }
        @Override
        public void merge(AggregationBuffer aggObject partialthrows HiveException {
            if (partial != null) {
                StdAgg myagg = (StdAggagg;
                Object partialCount = .getStructFieldData(partial);
                Object partialXAvg = .getStructFieldData(partial);
                Object partialYAvg = .getStructFieldData(partial);
                Object partialCovar = .getStructFieldData(partial);
                long nA = myagg.count;
                long nB = .get(partialCount);
                if (nA == 0) {
                    // Just copy the information since there is nothing so far
                    myagg.count = .get(partialCount);
                    myagg.xavg = .get(partialXAvg);
                    myagg.yavg = .get(partialYAvg);
                    myagg.covar = .get(partialCovar);
                }
                if (nA != 0 && nB != 0) {
                    // Merge the two partials
                    double xavgA = myagg.xavg;
                    double yavgA = myagg.yavg;
                    double xavgB = .get(partialXAvg);
                    double yavgB = .get(partialYAvg);
                    double covarB = .get(partialCovar);
                    myagg.count += nB;
                    myagg.xavg = (xavgA * nA + xavgB * nB) / myagg.count;
                    myagg.yavg = (yavgA * nA + yavgB * nB) / myagg.count;
                    myagg.covar += covarB + (xavgA - xavgB) * (yavgA - yavgB) * ((double) (nA * nB) / myagg.count);
                }
            }
        }
        @Override
        public Object terminate(AggregationBuffer aggthrows HiveException {
            StdAgg myagg = (StdAggagg;
            if (myagg.count == 0) { // SQL standard - return null for zero
                                    // elements
                return null;
            } else {
                getResult().set(myagg.covar / (myagg.count));
                return getResult();
            }
        }
        public void setResult(DoubleWritable result) {
            this. = result;
        }
        public DoubleWritable getResult() {
            return ;
        }
    }
New to GrepCode? Check out our FAQ X