hotime/vendor/github.com/montanaflynn/stats/data_set_distances.go
2022-10-19 21:32:34 +08:00

95 lines
2.5 KiB
Go

package stats
import (
"math"
)
// Validate data for distance calculation
func validateData(dataPointX, dataPointY []float64) error {
if len(dataPointX) == 0 || len(dataPointY) == 0 {
return EmptyInput
}
if len(dataPointX) != len(dataPointY) {
return SizeErr
}
return nil
}
// Computes Chebyshev distance between two data sets
func ChebyshevDistance(dataPointX, dataPointY []float64) (distance float64, err error) {
err = validateData(dataPointX, dataPointY)
if err != nil {
return math.NaN(), err
}
var tempDistance float64
for i := 0; i < len(dataPointY); i++ {
tempDistance = math.Abs(dataPointX[i] - dataPointY[i])
if distance < tempDistance {
distance = tempDistance
}
}
return distance, nil
}
//
// Computes Euclidean distance between two data sets
//
func EuclideanDistance(dataPointX, dataPointY []float64) (distance float64, err error) {
err = validateData(dataPointX, dataPointY)
if err != nil {
return math.NaN(), err
}
distance = 0
for i := 0; i < len(dataPointX); i++ {
distance = distance + ((dataPointX[i] - dataPointY[i]) * (dataPointX[i] - dataPointY[i]))
}
return math.Sqrt(distance), nil
}
//
// Computes Manhattan distance between two data sets
//
func ManhattanDistance(dataPointX, dataPointY []float64) (distance float64, err error) {
err = validateData(dataPointX, dataPointY)
if err != nil {
return math.NaN(), err
}
distance = 0
for i := 0; i < len(dataPointX); i++ {
distance = distance + math.Abs(dataPointX[i]-dataPointY[i])
}
return distance, nil
}
//
// Computes minkowski distance between two data sets.
//
// Input:
// dataPointX: First set of data points
// dataPointY: Second set of data points. Length of both data
// sets must be equal.
// lambda: aka p or city blocks; With lambda = 1
// returned distance is manhattan distance and
// lambda = 2; it is euclidean distance. Lambda
// reaching to infinite - distance would be chebysev
// distance.
// Output:
// Distance or error
//
func MinkowskiDistance(dataPointX, dataPointY []float64, lambda float64) (distance float64, err error) {
err = validateData(dataPointX, dataPointY)
if err != nil {
return math.NaN(), err
}
for i := 0; i < len(dataPointY); i++ {
distance = distance + math.Pow(math.Abs(dataPointX[i]-dataPointY[i]), lambda)
}
distance = math.Pow(distance, float64(1/lambda))
if math.IsInf(distance, 1) == true {
return math.NaN(), InfValue
}
return distance, nil
}