hbook is a set of data analysis tools for HEP (histograms (1D, 2D, 3D), profiles and ntuples).
hbook is a work in progress of a concurrent friendly histogram filling toolkit.
It is loosely based on AIDA interfaces and concepts as well as the "simplicity" of HBOOK and the previous work of YODA.
$ go get go-hep.org/x/hep/hbookDocumentation is available on godoc:
https://godoc.org/go-hep.org/x/hep/hbook
func ExampleH1D() {
const npoints = 10000
// Create a normal distribution.
dist := distuv.Normal{
Mu: 0,
Sigma: 1,
Src: rand.New(rand.NewSource(0)),
}
// Draw some random values from the standard
// normal distribution.
h := hbook.NewH1D(20, -4, +4)
for i := 0; i < npoints; i++ {
v := dist.Rand()
h.Fill(v, 1)
}
// fill h with a slice of values and their weights
h.FillN([]float64{1, 2, 3}, []float64{1, 1, 1})
h.FillN([]float64{1, 2, 3}, nil) // all weights are 1.
fmt.Printf("mean: %v\n", h.XMean())
fmt.Printf("rms: %v\n", h.XRMS())
fmt.Printf("std-dev: %v\n", h.XStdDev())
fmt.Printf("std-err: %v\n", h.XStdErr())
// Output:
// mean: 0.005589967511734562
// rms: 1.0062596231244403
// std-dev: 1.0062943821322063
// std-err: 0.010059926295994191
}func ExampleAddH1D() {
h1 := hbook.NewH1D(6, 0, 6)
h1.Fill(-0.5, 1)
h1.Fill(0, 1.5)
h1.Fill(0.5, 1)
h1.Fill(1.2, 1)
h1.Fill(2.1, 2)
h1.Fill(4.2, 1)
h1.Fill(5.9, 1)
h1.Fill(6, 0.5)
h2 := hbook.NewH1D(6, 0, 6)
h2.Fill(-0.5, 0.7)
h2.Fill(0.2, 1)
h2.Fill(0.7, 1.2)
h2.Fill(1.5, 0.8)
h2.Fill(2.2, 0.7)
h2.Fill(4.3, 1.3)
h2.Fill(5.2, 2)
h2.Fill(6.8, 1)
hsum := hbook.AddH1D(h1, h2)
fmt.Printf("Under: %.1f +/- %.1f\n", hsum.Binning.Outflows[0].SumW(), math.Sqrt(hsum.Binning.Outflows[0].SumW2()))
for i := 0; i < hsum.Len(); i++ {
fmt.Printf("Bin %v: %.1f +/- %.1f\n", i, hsum.Binning.Bins[i].SumW(), math.Sqrt(hsum.Binning.Bins[i].SumW2()))
}
fmt.Printf("Over : %.1f +/- %.1f\n", hsum.Binning.Outflows[1].SumW(), math.Sqrt(hsum.Binning.Outflows[1].SumW2()))
// Output:
// Under: 1.7 +/- 1.2
// Bin 0: 4.7 +/- 2.4
// Bin 1: 1.8 +/- 1.3
// Bin 2: 2.7 +/- 2.1
// Bin 3: 0.0 +/- 0.0
// Bin 4: 2.3 +/- 1.6
// Bin 5: 3.0 +/- 2.2
// Over : 1.5 +/- 1.1
}func ExampleAddScaledH1D() {
h1 := hbook.NewH1D(6, 0, 6)
h1.Fill(-0.5, 1)
h1.Fill(0, 1.5)
h1.Fill(0.5, 1)
h1.Fill(1.2, 1)
h1.Fill(2.1, 2)
h1.Fill(4.2, 1)
h1.Fill(5.9, 1)
h1.Fill(6, 0.5)
h2 := hbook.NewH1D(6, 0, 6)
h2.Fill(-0.5, 0.7)
h2.Fill(0.2, 1)
h2.Fill(0.7, 1.2)
h2.Fill(1.5, 0.8)
h2.Fill(2.2, 0.7)
h2.Fill(4.3, 1.3)
h2.Fill(5.2, 2)
h2.Fill(6.8, 1)
hsum := hbook.AddScaledH1D(h1, 10, h2)
fmt.Printf("Under: %.1f +/- %.1f\n", hsum.Binning.Outflows[0].SumW(), math.Sqrt(hsum.Binning.Outflows[0].SumW2()))
for i := 0; i < hsum.Len(); i++ {
fmt.Printf("Bin %v: %.1f +/- %.1f\n", i, hsum.Binning.Bins[i].SumW(), math.Sqrt(hsum.Binning.Bins[i].SumW2()))
}
fmt.Printf("Over : %.1f +/- %.1f\n", hsum.Binning.Outflows[1].SumW(), math.Sqrt(hsum.Binning.Outflows[1].SumW2()))
// Output:
// Under: 8.0 +/- 7.1
// Bin 0: 24.5 +/- 15.7
// Bin 1: 9.0 +/- 8.1
// Bin 2: 9.0 +/- 7.3
// Bin 3: 0.0 +/- 0.0
// Bin 4: 14.0 +/- 13.0
// Bin 5: 21.0 +/- 20.0
// Over : 10.5 +/- 10.0
}func ExampleSubH1D() {
h1 := hbook.NewH1D(6, 0, 6)
h1.Fill(-0.5, 1)
h1.Fill(0, 1.5)
h1.Fill(0.5, 1)
h1.Fill(1.2, 1)
h1.Fill(2.1, 2)
h1.Fill(4.2, 1)
h1.Fill(5.9, 1)
h1.Fill(6, 0.5)
h2 := hbook.NewH1D(6, 0, 6)
h2.Fill(-0.5, 0.7)
h2.Fill(0.2, 1)
h2.Fill(0.7, 1.2)
h2.Fill(1.5, 0.8)
h2.Fill(2.2, 0.7)
h2.Fill(4.3, 1.3)
h2.Fill(5.2, 2)
h2.Fill(6.8, 1)
hsub := hbook.SubH1D(h1, h2)
under := hsub.Binning.Outflows[0]
fmt.Printf("Under: %.1f +/- %.1f\n", under.SumW(), math.Sqrt(under.SumW2()))
for i, bin := range hsub.Binning.Bins {
fmt.Printf("Bin %v: %.1f +/- %.1f\n", i, bin.SumW(), math.Sqrt(bin.SumW2()))
}
over := hsub.Binning.Outflows[1]
fmt.Printf("Over : %.1f +/- %.1f\n", over.SumW(), math.Sqrt(over.SumW2()))
// Output:
// Under: 0.3 +/- 1.2
// Bin 0: 0.3 +/- 2.4
// Bin 1: 0.2 +/- 1.3
// Bin 2: 1.3 +/- 2.1
// Bin 3: 0.0 +/- 0.0
// Bin 4: -0.3 +/- 1.6
// Bin 5: -1.0 +/- 2.2
// Over : -0.5 +/- 1.1
}func ExampleH2D() {
h := hbook.NewH2D(100, -10, 10, 100, -10, 10)
const npoints = 10000
dist, ok := distmv.NewNormal(
[]float64{0, 1},
mat.NewSymDense(2, []float64{4, 0, 0, 2}),
rand.New(rand.NewSource(1234)),
)
if !ok {
log.Fatalf("error creating distmv.Normal")
}
v := make([]float64, 2)
// Draw some random values from the standard
// normal distribution.
for i := 0; i < npoints; i++ {
v = dist.Rand(v)
h.Fill(v[0], v[1], 1)
}
// fill h with slices of values and their weights
h.FillN(
[]float64{1, 2, 3}, // xs
[]float64{1, 2, 3}, // ys
[]float64{1, 1, 1}, // ws
)
// fill h with slices of values. all weights are 1.
h.FillN(
[]float64{1, 2, 3}, // xs
[]float64{1, 2, 3}, // ys
nil, // ws
)
}func ExampleS2D() {
s := hbook.NewS2D(hbook.Point2D{X: 1, Y: 1}, hbook.Point2D{X: 2, Y: 1.5}, hbook.Point2D{X: -1, Y: +2})
if s == nil {
log.Fatal("nil pointer to S2D")
}
fmt.Printf("len=%d\n", s.Len())
s.Fill(hbook.Point2D{X: 10, Y: -10, ErrX: hbook.Range{Min: 5, Max: 5}, ErrY: hbook.Range{Min: 6, Max: 6}})
fmt.Printf("len=%d\n", s.Len())
fmt.Printf("pt[%d]=%+v\n", 3, s.Point(3))
// Output:
// len=3
// len=4
// pt[3]={X:10 Y:-10 ErrX:{Min:5 Max:5} ErrY:{Min:6 Max:6}}
}package main
import (
"database/sql"
"fmt"
_ "go-hep.org/x/hep/csvutil/csvdriver"
"go-hep.org/x/hep/hbook/ntup"
)
func main() {
db, err := sql.Open("csv", "data.csv")
if err != nil {
panic(err)
}
defer db.Close()
nt, err := ntup.Open(db, "csv")
if err != nil {
panic(err)
}
h1, err := nt.ScanH1D("px where pt>100", nil)
if err != nil {
panic(err)
}
fmt.Printf("h1: %v\n", h1)
h2 := hbook.NewH1D(100, -10, 10)
h2, err = nt.ScanH1D("px where pt>100 && pt < 1000", h2)
if err != nil {
panic(err)
}
fmt.Printf("h2: %v\n", h2)
h11 := hbook.NewH1D(100, -10, 10)
h22 := hbook.NewH1D(100, -10, 10)
err = nt.Scan("px, py where pt>100", func(px, py float64) error {
h11.Fill(px, 1)
h22.Fill(py, 1)
return nil
})
if err != nil {
panic(err)
}
fmt.Printf("h11: %v\n", h11)
fmt.Printf("h22: %v\n", h22)
}