README
¶
gostats
Statistics for golang
Usage
To install, do go get github.com/r0fls/gostats. Check out gostats_test.go for a working example of using each distribution.
Example
package main
import (
"fmt"
"github.com/r0fls/gostats"
)
func main() {
b := stats.Bernoulli(.5)
fmt.Println(b.Random())
// or fit a distribution from data...
b = stats.FitBernoulli([0,1,1,1])
fmt.Println(b.Random())
}
Distributions
Currently the following distributions are included:
- Bernoulli
- Laplace
- Poisson
- Geometric
- Weibull
- Exponential
- Binomial
- NegativeBinomial *
*note the negative binomial takes parameters
r,pwhereris the number allowed success before termination of the trials andpis the success of a given trial. The random variable produced byNegativeBinomialis the number of failures.
And each distribution has these functions:
- Pmf or Pdf
- Cdf
- Quantile
- Random
Also there is a corresponding function named FitDistrbution for each distribution, as shown in the above example with the Bernoulli. That function uses the MLE for each distribution to choose the best estimation for the parameters and returns an initialized distribution with them.
Common
Additionally there are some common functions. Most notably is LSR, which performs a least squares regression.
Additional Info
The default seed function is time.Now().UTC().UnixNano(). However you can override that by calling rand.Seed(uniqueInteger) yourself before generating random numbers.
Contributing
If you're interested in contributing, please submit a pull request, or raise an issue.
- add more distributions
- should
randomreturn an array? - allow updating a fitted distribution with more data
Documentation
¶
Index ¶
- func Choose(n int, k int) int
- func Factorial(n int) int
- func LSR(data [][]float64) []float64
- func MeanFloat64(data []float64) float64
- func MeanInt(data []int) float64
- func MedianFloat64(data []float64) float64
- func MedianInt(data []int) float64
- func Seed()
- func SumFloat64(data []float64) float64
- func SumInt(data []int) int
- type BernoulliType
- type BinomialType
- type ExponentialType
- type GeometricType
- type LaplaceType
- type NegativeBinomialType
- type PoissonType
- type WeibullType
Examples ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func LSR ¶
Example ¶
data := [][]float64{{60.0, 3.1}, {61.0, 3.6}, {62.0, 3.8}, {63, 4}, {65.0, 4.1}}
fmt.Println(stats.LSR(data))
Output: [-7.963513513513208 0.18783783783783292]
func MeanFloat64 ¶
func MedianFloat64 ¶
func SumFloat64 ¶
Example ¶
s := []float64{2, 3, 5, 7, 11, 13}
sum := stats.SumFloat64(s)
fmt.Println(sum)
Output: 41
Types ¶
type BernoulliType ¶
type BernoulliType struct {
// contains filtered or unexported fields
}
func Bernoulli ¶
func Bernoulli(p float64) BernoulliType
func FitBernoulli ¶
func FitBernoulli(data []int) BernoulliType
Example ¶
data := []int{1, 1, 1, 0, 0}
b := stats.FitBernoulli(data)
fmt.Println(b.Cdf(0))
Output: 0.4
func (BernoulliType) Quantile ¶
func (b BernoulliType) Quantile(p float64) int
type BinomialType ¶
type BinomialType struct {
// contains filtered or unexported fields
}
func Binomial ¶
func Binomial(n int, p float64) BinomialType
func (BinomialType) Quantile ¶
func (b BinomialType) Quantile(x float64) int
type ExponentialType ¶
type ExponentialType struct {
// contains filtered or unexported fields
}
func Exponential ¶
func Exponential(l float64) ExponentialType
func FitExponential ¶
func FitExponential(data []float64) ExponentialType
Example ¶
data := []float64{10, 3, 3, 4, 5}
g := stats.FitExponential(data)
fmt.Println(g.Quantile(.5))
Output: 3.465735902799726
func (ExponentialType) Quantile ¶
func (e ExponentialType) Quantile(p float64) float64
type GeometricType ¶
type GeometricType struct {
// contains filtered or unexported fields
}
func FitGeometric ¶
func FitGeometric(data []int) GeometricType
Example ¶
data := []int{10, 3, 3, 4, 5}
g := stats.FitGeometric(data)
fmt.Println(g.Quantile(.5))
Output: 4
func Geometric ¶
func Geometric(p float64) GeometricType
func (GeometricType) Quantile ¶
func (g GeometricType) Quantile(p float64) int
type LaplaceType ¶
type LaplaceType struct {
// contains filtered or unexported fields
}
func FitLaplace ¶
func FitLaplace(data []float64) LaplaceType
Example ¶
data := []float64{1, 1, 1, 0, 0}
l := stats.FitLaplace(data)
fmt.Println(l.Quantile(.5))
Output: 1
func Laplace ¶
func Laplace(mean float64, b float64) LaplaceType
func (LaplaceType) Quantile ¶
func (l LaplaceType) Quantile(p float64) float64
type NegativeBinomialType ¶
type NegativeBinomialType struct {
// contains filtered or unexported fields
}
func NegativeBinomial ¶
func NegativeBinomial(k int, p float64) NegativeBinomialType
func (NegativeBinomialType) Quantile ¶
func (b NegativeBinomialType) Quantile(x float64) int
type PoissonType ¶
type PoissonType struct {
// contains filtered or unexported fields
}
func FitPoisson ¶
func FitPoisson(data []int) PoissonType
Example ¶
data := []int{10, 3, 3, 4, 5}
p := stats.FitPoisson(data)
fmt.Println(p.Quantile(.5))
Output: 5
func Poisson ¶
func Poisson(m float64) PoissonType
func (PoissonType) Quantile ¶
func (p PoissonType) Quantile(x float64) int
type WeibullType ¶
type WeibullType struct {
// contains filtered or unexported fields
}
func Weibull ¶
func Weibull(l float64, k float64) WeibullType
func (WeibullType) Quantile ¶
func (w WeibullType) Quantile(p float64) float64
Source Files