README
¶
Gonum floats 
Package floats provides a set of helper routines for dealing with slices of float64. The functions avoid allocations to allow for use within tight loops without garbage collection overhead.
Documentation
¶
Overview ¶
Package floats provides a set of helper routines for dealing with slices of float64. The functions avoid allocations to allow for use within tight loops without garbage collection overhead.
The convention used is that when a slice is being modified in place, it has the name dst.
Index ¶
- func Add(dst, s []float64)
- func AddConst(c float64, dst []float64)
- func AddScaled(dst []float64, alpha float64, s []float64)
- func AddScaledTo(dst, y []float64, alpha float64, s []float64) []float64
- func AddTo(dst, s, t []float64) []float64
- func Argsort(dst []float64, inds []int)
- func Count(f func(float64) bool, s []float64) int
- func CumProd(dst, s []float64) []float64
- func CumSum(dst, s []float64) []float64
- func Distance(s, t []float64, L float64) float64
- func Div(dst, s []float64)
- func DivTo(dst, s, t []float64) []float64
- func Dot(s1, s2 []float64) float64
- func Equal(s1, s2 []float64) bool
- func EqualApprox(s1, s2 []float64, tol float64) bool
- func EqualFunc(s1, s2 []float64, f func(float64, float64) bool) bool
- func EqualLengths(slices ...[]float64) bool
- func Find(inds []int, f func(float64) bool, s []float64, k int) ([]int, error)
- func HasNaN(s []float64) bool
- func LogSpan(dst []float64, l, u float64) []float64
- func LogSumExp(s []float64) float64
- func Max(s []float64) float64
- func MaxIdx(s []float64) int
- func Min(s []float64) float64
- func MinIdx(s []float64) int
- func Mul(dst, s []float64)
- func MulTo(dst, s, t []float64) []float64
- func NearestIdx(s []float64, v float64) int
- func NearestIdxForSpan(n int, l, u float64, v float64) int
- func Norm(s []float64, L float64) float64
- func Prod(s []float64) float64
- func Reverse(s []float64)
- func Same(s, t []float64) bool
- func Scale(c float64, dst []float64)
- func ScaleTo(dst []float64, c float64, s []float64) []float64
- func Span(dst []float64, l, u float64) []float64
- func Sub(dst, s []float64)
- func SubTo(dst, s, t []float64) []float64
- func Sum(s []float64) float64
- func SumCompensated(s []float64) float64
- func Within(s []float64, v float64) int
Examples ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func Add ¶
func Add(dst, s []float64)
Add adds, element-wise, the elements of s and dst, and stores the result in dst. It panics if the argument lengths do not match.
Example (Newslice) ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
// If one wants to store the result in a
// new container, just make a new slice
s1 := []float64{1, 2, 3, 4}
s2 := []float64{5, 6, 7, 8}
s3 := []float64{1, 1, 1, 1}
dst := make([]float64, len(s1))
floats.AddTo(dst, s1, s2)
floats.Add(dst, s3)
fmt.Println("dst =", dst)
fmt.Println("s1 =", s1)
fmt.Println("s2 =", s2)
fmt.Println("s3 =", s3)
}
Output: dst = [7 9 11 13] s1 = [1 2 3 4] s2 = [5 6 7 8] s3 = [1 1 1 1]
Example (Simple) ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
// Adding three slices together. Note that
// the result is stored in the first slice
s1 := []float64{1, 2, 3, 4}
s2 := []float64{5, 6, 7, 8}
s3 := []float64{1, 1, 1, 1}
floats.Add(s1, s2)
floats.Add(s1, s3)
fmt.Println("s1 =", s1)
fmt.Println("s2 =", s2)
fmt.Println("s3 =", s3)
}
Output: s1 = [7 9 11 13] s2 = [5 6 7 8] s3 = [1 1 1 1]
Example (Unequallengths) ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
// If the lengths of the slices are unknown,
// use EqualLengths to check
s1 := []float64{1, 2, 3}
s2 := []float64{5, 6, 7, 8}
eq := floats.EqualLengths(s1, s2)
if eq {
floats.Add(s1, s2)
} else {
fmt.Println("Unequal lengths")
}
}
Output: Unequal lengths
func AddConst ¶
AddConst adds the scalar c to all of the values in dst.
Example ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
s := []float64{1, -2, 3, -4}
c := 5.0
floats.AddConst(c, s)
fmt.Println("s =", s)
}
Output: s = [6 3 8 1]
func AddScaled ¶
AddScaled performs dst = dst + alpha * s. It panics if the slice argument lengths do not match.
func AddScaledTo ¶
AddScaledTo performs dst = y + alpha * s, where alpha is a scalar, and dst, y and s are all slices. It panics if the slice argument lengths do not match.
At the return of the function, dst[i] = y[i] + alpha * s[i]
func AddTo ¶
AddTo adds, element-wise, the elements of s and t and stores the result in dst. It panics if the argument lengths do not match.
func Argsort ¶
Argsort sorts the elements of dst while tracking their original order. At the conclusion of Argsort, dst will contain the original elements of dst but sorted in increasing order, and inds will contain the original position of the elements in the slice such that dst[i] = origDst[inds[i]]. It panics if the argument lengths do not match.
func Count ¶
Count applies the function f to every element of s and returns the number of times the function returned true.
func CumProd ¶
CumProd finds the cumulative product of the first i elements in s and puts them in place into the ith element of the destination dst. It panics if the argument lengths do not match.
At the return of the function, dst[i] = s[i] * s[i-1] * s[i-2] * ...
Example ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
s := []float64{1, -2, 3, -4}
dst := make([]float64, len(s))
floats.CumProd(dst, s)
fmt.Println("dst =", dst)
fmt.Println("s =", s)
}
Output: dst = [1 -2 -6 24] s = [1 -2 3 -4]
func CumSum ¶
CumSum finds the cumulative sum of the first i elements in s and puts them in place into the ith element of the destination dst. It panics if the argument lengths do not match.
At the return of the function, dst[i] = s[i] + s[i-1] + s[i-2] + ...
Example ¶
package main
import (
"fmt"
"gonum.org/v1/gonum/floats"
)
func main() {
s := []float64{1, -2, 3, -4}
dst := make([]float64, len(s))
floats.CumSum(dst, s)
fmt.Println("dst =", dst)
fmt.Println("s =", s)
}
Output: dst = [1 -1 2 -2] s = [1 -2 3 -4]
func Distance ¶
Distance computes the L-norm of s - t. See Norm for special cases. It panics if the slice argument lengths do not match.
func Div ¶
func Div(dst, s []float64)
Div performs element-wise division dst / s and stores the value in dst. It panics if the argument lengths do not match.
func DivTo ¶
DivTo performs element-wise division s / t and stores the value in dst. It panics if the argument lengths do not match.
func Dot ¶
Dot computes the dot product of s1 and s2, i.e. sum_{i = 1}^N s1[i]*s2[i]. It panics if the argument lengths do not match.
func Equal ¶
Equal returns true when the slices have equal lengths and all elements are numerically identical.
func EqualApprox ¶
EqualApprox returns true when the slices have equal lengths and all element pairs have an absolute tolerance less than tol or a relative tolerance less than tol.
func EqualFunc ¶
EqualFunc returns true when the slices have the same lengths and the function returns true for all element pairs.
func EqualLengths ¶
EqualLengths returns true when all of the slices have equal length, and false otherwise. It also returns true when there are no input slices.
func Find ¶
Find applies f to every element of s and returns the indices of the first k elements for which the f returns true, or all such elements if k < 0. Find will reslice inds to have 0 length, and will append found indices to inds. If k > 0 and there are fewer than k elements in s satisfying f, all of the found elements will be returned along with an error. At the return of the function, the input inds will be in an undetermined state.
func LogSpan ¶
LogSpan returns a set of n equally spaced points in log space between, l and u where N is equal to len(dst). The first element of the resulting dst will be l and the final element of dst will be u. It panics if the length of dst is less than 2. Note that this call will return NaNs if either l or u are negative, and will return all zeros if l or u is zero. Also returns the mutated slice dst, so that it can be used in range, like:
for i, x := range LogSpan(dst, l, u) { ... }
func LogSumExp ¶
LogSumExp returns the log of the sum of the exponentials of the values in s. Panics if s is an empty slice.
func MaxIdx ¶
MaxIdx returns the index of the maximum value in the input slice. If several entries have the maximum value, the first such index is returned. It panics if s is zero length.
func MinIdx ¶
MinIdx returns the index of the minimum value in the input slice. If several entries have the minimum value, the first such index is returned. It panics if s is zero length.
func Mul ¶
func Mul(dst, s []float64)
Mul performs element-wise multiplication between dst and s and stores the value in dst. It panics if the argument lengths do not match.
func MulTo ¶
MulTo performs element-wise multiplication between s and t and stores the value in dst. It panics if the argument lengths do not match.
func NearestIdx ¶
NearestIdx returns the index of the element in s whose value is nearest to v. If several such elements exist, the lowest index is returned. It panics if s is zero length.
func NearestIdxForSpan ¶
NearestIdxForSpan return the index of a hypothetical vector created by Span with length n and bounds l and u whose value is closest to v. That is, NearestIdxForSpan(n, l, u, v) is equivalent to Nearest(Span(make([]float64, n),l,u),v) without an allocation. It panics if n is less than two.
func Norm ¶
Norm returns the L norm of the slice S, defined as (sum_{i=1}^N s[i]^L)^{1/L} Special cases: L = math.Inf(1) gives the maximum absolute value. Does not correctly compute the zero norm (use Count).
func Same ¶
Same returns true when the input slices have the same length and all elements have the same value with NaN treated as the same.
func ScaleTo ¶
ScaleTo multiplies the elements in s by c and stores the result in dst. It panics if the slice argument lengths do not match.
func Span ¶
Span returns a set of N equally spaced points between l and u, where N is equal to the length of the destination. The first element of the destination is l, the final element of the destination is u. It panics if the length of dst is less than 2.
Span also returns the mutated slice dst, so that it can be used in range expressions, like:
for i, x := range Span(dst, l, u) { ... }
func Sub ¶
func Sub(dst, s []float64)
Sub subtracts, element-wise, the elements of s from dst. It panics if the argument lengths do not match.
func SubTo ¶
SubTo subtracts, element-wise, the elements of t from s and stores the result in dst. It panics if the argument lengths do not match.
func SumCompensated ¶
SumCompensated returns the sum of the elements of the slice calculated with greater accuracy than Sum at the expense of additional computation.
Types ¶
This section is empty.
Source Files
Directories