mathrock

A small math library for Go, providing Vector and Matrix types with common linear algebra operations.

---

Installation

go get github.com/areyoush/mathrock

---

Vector

Import: "github.com/areyoush/mathrock/vector"

Vector is defined as type Vector []float64.

Constructing

a := vector.Vector{1, 2, 3}
b := vector.Vector{4, 5, 6}

Methods

Arithmetic & Scaling

Method	Call	Behavior on invalid input
Add	a.Add(b)	panics on length mismatch
Subtract	a.Subtract(b)	panics on length mismatch
Multiply	a.Multiply(b)	panics on length mismatch
Scale	a.Scale(2)	never fails

Spatial & Statistical Operations

Method	Call	Behavior on invalid input
Dot	a.Dot(b)	panics on length mismatch
Norm	a.Norm()	never fails
Normalize	a.Normalize()	returns error for the zero vector
Distance	a.Distance(b)	panics on length mismatch
Sum	a.Sum()	never fails
Mean	a.Mean()	returns error for an empty vector

Utility Methods

Method	Call	Behavior on invalid input
Equals	a.Equals(b)	returns false on length mismatch
EqualsWithTolerance	a.EqualsWithTolerance(b, 1e-6)	returns false on length mismatch

Examples

a := vector.Vector{1, 2, 3}
b := vector.Vector{4, 5, 6}

a.Dot(b)                     // 32
a.Add(b)                     // [5 7 9]
a.Subtract(b)                // [-3 -3 -3]
a.Scale(2)                   // [2 4 6]
a.Multiply(b)                // [4 10 18]
a.Norm()                     // 3.7416573867739413
a.Distance(b)                // 5.196152422706632

normalized, err := a.Normalize()
if err != nil {
    // v was the zero vector
}

mean, err := a.Mean()
if err != nil {
    // v was empty
}

a.Equals(a)                  // true
a.EqualsWithTolerance(b, 10) // true (loose tolerance)

---

Matrix

Import: "github.com/areyoush/mathrock/matrix"

Matrix stores its values internally as a flat []float64 in row-major order.

Constructing

m, err := matrix.NewMatrix(2, 3, []float64{1, 2, 3, 4, 5, 6})
// m:
// [1 2 3]
// [4 5 6]

NewMatrix returns an error if len(data) != rows*cols.

Methods

Initialization & Access

Method	Call	Behavior on invalid input
At	m.At(1, 2)	panics on out-of-bounds index
Set	m.Set(0, 0, 99)	panics on out-of-bounds index
Row	m.Row(1)	panics on out-of-bounds index
Column	m.Column(1)	panics on out-of-bounds index

Mathematical Operations

Method	Call	Behavior on invalid input
Add	m.Add(other)	panics on dimension mismatch
Subtract	m.Subtract(other)	panics on dimension mismatch
Scale	m.Scale(2)	never fails
Matmul	m.Matmul(other)	panics if m.cols != other.rows
Transpose / T	m.Transpose() / m.T()	never fails

Utility Methods

Method	Call	Behavior on invalid input
Rows	m.Rows()	never fails
Cols	m.Cols()	never fails
Equals	m.Equals(other)	returns false on dimension mismatch
EqualsWithTolerance	m.EqualsWithTolerance(other, 1e-6)	returns false on dimension mismatch
String	fmt.Println(m)	never fails

Examples

a, _ := matrix.NewMatrix(2, 3, []float64{1, 2, 3, 4, 5, 6})
b, _ := matrix.NewMatrix(3, 2, []float64{7, 8, 9, 10, 11, 12})

a.At(1, 2)           // 6
a.Set(0, 0, 99)      // mutates a in place
a.Row(1)             // [4 5 6]
a.Column(1)          // [2 5]

a.Scale(2)
// [2 4 6]
// [8 10 12]

a.Matmul(b)
// [58 64]
// [139 154]

a.Transpose()
// [1 4]
// [2 5]
// [3 6]

a.Rows()     // 2
a.Cols()     // 3

a.Equals(a)  // true

Printing a matrix directly (fmt.Println(m)) prints it row-by-row rather than as a flat slice, since Matrix implements fmt.Stringer.

---

Status

This is an early, actively developed library (v0.1.x). The API may still change as more methods are added. Contributions and suggestions are welcome.