NumSharp API Reference

NumSharp is a .NET port of Python's NumPy library. This API reference is organized by functionality, matching NumPy's documentation structure.

Core Types

The essential types for working with NumSharp.

Type	Description
NDArray	The main n-dimensional array type
np	Static API class (like `import numpy as np` in Python)
Shape	Array dimensions and strides
Slice	Slice specification for array indexing
NDArray<TDType>	Generic typed wrapper for type-safe access

Quick Example

using NumSharp;

// Create arrays
var a = np.array(new[] { 1, 2, 3, 4, 5 });
var b = np.zeros((3, 4));
var c = np.arange(10);

// Operations
var sum = np.sum(a);
var reshaped = a.reshape(5, 1);
var sliced = a["1:4"];  // Elements 1, 2, 3

Array Creation

Functions for creating new arrays.

Function	Description
`np.array(data)`	Create array from existing data
`np.zeros(shape)`	Array filled with zeros
`np.zeros_like(a)`	Array of zeros with same shape as `a`
`np.ones(shape)`	Array filled with ones
`np.ones_like(a)`	Array of ones with same shape as `a`
`np.empty(shape)`	Uninitialized array
`np.full(shape, value)`	Array filled with a constant value
`np.eye(N)`	Identity matrix
`np.arange(start, stop, step)`	Evenly spaced values within interval
`np.linspace(start, stop, num)`	Evenly spaced values (specify count)
`np.meshgrid(x, y)`	Coordinate matrices from vectors
`np.copy(a)`	Return a copy of the array
`np.asarray(a)`	Convert input to array
`np.frombuffer(buffer)`	Create array from buffer

Stacking & Joining

Functions for combining multiple arrays.

Function	Description
`np.concatenate(arrays, axis)`	Join arrays along an existing axis
`np.stack(arrays, axis)`	Join arrays along a new axis
`np.vstack(arrays)`	Stack arrays vertically (row-wise)
`np.hstack(arrays)`	Stack arrays horizontally (column-wise)
`np.dstack(arrays)`	Stack arrays depth-wise (along 3rd axis)

Math Operations

Arithmetic and mathematical functions.

Arithmetic Operators

Operator	Description
`a + b`	Element-wise addition
`a - b`	Element-wise subtraction
`a * b`	Element-wise multiplication
`a / b`	Element-wise division
`a % b`	Element-wise modulo
`-a`	Element-wise negation

Math Functions

Function	Description
`np.sum(a, axis)`	Sum of array elements
`np.prod(a)`	Product of array elements
`np.cumsum(a)`	Cumulative sum
`np.sqrt(a)`	Element-wise square root
`np.power(a, n)`	Element-wise power
`np.abs(a)`	Element-wise absolute value
`np.sign(a)`	Element-wise sign
`np.floor(a)`	Element-wise floor
`np.ceil(a)`	Element-wise ceiling
`np.round(a)`	Element-wise rounding
`np.clip(a, min, max)`	Clip values to range
`np.maximum(a, b)`	Element-wise maximum
`np.minimum(a, b)`	Element-wise minimum

Exponentials & Logarithms

Function	Description
`np.exp(a)`	Element-wise exponential
`np.exp2(a)`	Element-wise 2^x
`np.expm1(a)`	exp(x) - 1
`np.log(a)`	Natural logarithm
`np.log2(a)`	Base-2 logarithm
`np.log10(a)`	Base-10 logarithm
`np.log1p(a)`	log(1 + x)

Trigonometric Functions

Function	Description
`np.sin(a)`	Element-wise sine
`np.cos(a)`	Element-wise cosine
`np.tan(a)`	Element-wise tangent

Statistics

Statistical functions.

Function	Description
`np.mean(a, axis)`	Arithmetic mean
`np.std(a, axis)`	Standard deviation
`np.var(a, axis)`	Variance
`np.amax(a, axis)`	Maximum value
`np.amin(a, axis)`	Minimum value

Sorting & Searching

Functions for sorting arrays and finding elements.

Function	Description
`np.argsort(a)`	Indices that would sort an array
`np.argmax(a, axis)`	Index of maximum value
`np.argmin(a, axis)`	Index of minimum value
`np.searchsorted(a, v)`	Find indices for inserting values
`np.nonzero(a)`	Indices of non-zero elements

Linear Algebra

Matrix and vector operations.

Function	Description
`np.dot(a, b)`	Dot product / matrix multiplication
`np.matmul(a, b)`	Matrix product (@ operator)
`np.outer(a, b)`	Outer product of two vectors

Shape Manipulation

Functions for changing array shape and dimensions.

Function	Description
`np.reshape(a, shape)`	Reshape without changing data
`a.reshape(shape)`	Instance method for reshape
`np.transpose(a)`	Permute array dimensions
`a.T`	Transpose property
`np.ravel(a)`	Flatten to 1-D array (returns view)
`a.flatten()`	Flatten to 1-D array (returns copy)
`np.squeeze(a)`	Remove axes of length 1
`np.expand_dims(a, axis)`	Insert a new axis
`np.swapaxes(a, ax1, ax2)`	Swap two axes
`np.moveaxis(a, src, dst)`	Move axes to new positions
`np.rollaxis(a, axis)`	Roll axis backwards
`np.atleast_1d(a)`	Convert to at least 1-D
`np.atleast_2d(a)`	Convert to at least 2-D
`np.atleast_3d(a)`	Convert to at least 3-D

Indexing & Slicing

NumSharp supports Python-style array indexing and slicing.

Slice Syntax

a[":"]           // All elements
a["1:5"]         // Elements 1-4 (stop exclusive)
a["::2"]         // Every 2nd element
a["-1"]          // Last element (reduces dimension)
a["::-1"]        // Reversed
a[":, 0"]        // All rows, first column
a["..., -1"]     // Ellipsis fills dimensions

Special Slice Constants

Constant	Description
`Slice.All`	All elements (`:`)
`Slice.Ellipsis`	Fill remaining dimensions (`...`)
`Slice.NewAxis`	Insert new dimension
`Slice.Index(n)`	Single element selection

Boolean Masking

var a = np.array(new[] { 1, 2, 3, 4, 5 });
var mask = a > 2;        // [false, false, true, true, true]
var filtered = a[mask];  // [3, 4, 5]

Logic Functions

Boolean operations and comparisons.

Comparison Operators

Operator	Description
`a == b`	Element-wise equality
`a != b`	Element-wise inequality
`a > b`	Element-wise greater than
`a >= b`	Element-wise greater or equal
`a < b`	Element-wise less than
`a <= b`	Element-wise less or equal
`!a`	Element-wise NOT (boolean arrays)

Logic Functions

Function	Description
`np.all(a, axis)`	Test if all elements are true
`np.any(a, axis)`	Test if any element is true
`np.array_equal(a, b)`	Test if arrays are equal
`np.isscalar(a)`	Test if input is scalar

Random Sampling

Random number generation. Access via NumPyRandom.

Function	Description
`np.random.rand(d0, d1, ...)`	Random values in [0, 1)
`np.random.randn(d0, d1, ...)`	Standard normal distribution
`np.random.randint(low, high, size)`	Random integers
`np.random.uniform(low, high, size)`	Uniform distribution
`np.random.choice(a, size, replace)`	Random sample from array
`np.random.shuffle(a)`	Shuffle array in-place
`np.random.permutation(a)`	Random permutation

Distributions

Function	Description
`np.random.beta(a, b, size)`	Beta distribution
`np.random.binomial(n, p, size)`	Binomial distribution
`np.random.gamma(shape, scale, size)`	Gamma distribution
`np.random.poisson(lam, size)`	Poisson distribution
`np.random.exponential(scale, size)`	Exponential distribution
`np.random.geometric(p, size)`	Geometric distribution
`np.random.lognormal(mean, sigma, size)`	Log-normal distribution
`np.random.chisquare(df, size)`	Chi-square distribution
`np.random.bernoulli(p, size)`	Bernoulli distribution

Broadcasting

Functions for array broadcasting.

Function	Description
`np.broadcast_to(a, shape)`	Broadcast array to new shape
`np.broadcast_arrays(a, b, ...)`	Broadcast arrays against each other

Broadcasting automatically aligns array shapes for operations:

Shapes align from the right
Dimensions must be equal OR one must be 1
Dimension of 1 "stretches" to match

File I/O

Functions for saving and loading arrays.

Function	Description
`np.save(file, arr)`	Save array to `.npy` file
`np.load(file)`	Load `.npy` or `.npz` file
`np.fromfile(file, dtype)`	Load array from binary file
`arr.tofile(file)`	Write array to binary file

Unique & Set Operations

Function	Description
`np.unique(a)`	Find unique elements
`np.repeat(a, repeats)`	Repeat elements of array

Internals & Advanced

These types are for advanced users extending NumSharp or understanding its internals.

Storage & Backends

Type	Description
UnmanagedStorage	Raw unmanaged memory management
TensorEngine	Abstract computation backend interface
DefaultEngine	Pure C# implementation of TensorEngine

Iteration

Type	Description
NDIterator	Traverses arrays with different memory layouts
MultiIterator	Paired iteration for broadcasting

Memory Management

Type	Description
ArraySlice<T>	Typed memory slice
IMemoryBlock	Memory block interface

Utilities

Type	Description
InfoOf<T>	Static type information cache
NPTypeCode	Enum of supported data types

Supported Data Types

NumSharp supports 12 numeric data types:

NPTypeCode	C# Type	NumPy Equivalent
Boolean	`bool`	`np.bool_`
Byte	`byte`	`np.uint8`
Int16	`short`	`np.int16`
UInt16	`ushort`	`np.uint16`
Int32	`int`	`np.int32`
UInt32	`uint`	`np.uint32`
Int64	`long`	`np.int64`
UInt64	`ulong`	`np.uint64`
Char	`char`	(no equivalent)
Single	`float`	`np.float32`
Double	`double`	`np.float64`
Decimal	`decimal`	(no equivalent)

Table of Contents

NumSharp API Reference

Core Types

Quick Example

Array Creation

Stacking & Joining

Math Operations

Arithmetic Operators

Math Functions

Exponentials & Logarithms

Trigonometric Functions

Statistics

Sorting & Searching

Linear Algebra

Shape Manipulation

Indexing & Slicing

Slice Syntax

Special Slice Constants

Boolean Masking

Logic Functions

Comparison Operators

Logic Functions

Random Sampling

Distributions

Broadcasting

File I/O

Unique & Set Operations

Internals & Advanced

Storage & Backends

Iteration

Memory Management

Utilities

Supported Data Types

See Also