GitHub - mlcpp/Matrix: Easy-to-use Scientific Computing library in/for C++ available for Linux and Windows. (original) (raw)

Matrix

Table of Contents

1. Installation
2. Development
   2.1. Linux
   2.2. Windows
3. Benchmarking
4. Testing
5. Quick Start Guide
   5.1. Initializers
   5.2. Slicing
   5.3. Printing/Viewing
   5.4. Indexing
   5.5. Operators
   5.6. Broadcasting
   5.7. Minimum, Maximum
   5.8. Mathematical Operations
   5.9. Statistical Operations
   5.10. Matrix Algebra
   5.11. Miscellaneous

Installation

This describes the installation process using cmake. As pre-requisites, you'll need git and cmake installed.

Check out the library.

$ git clone https://github.com/mlcpp/Matrix.git && cd Matrix

Make a build directory to place the build output.

$ cmake -E make_directory "build"

Generate build system files with cmake.

$ cmake -E chdir "build" cmake -DCMAKE_BUILD_TYPE=Release ../

Build the library.

$ cmake -DCMAKE_BUILD_TYPE=Release -S . -B "build"

On a unix system, the build directory should now look something like this:

/Matrix
  /build
    /examples
      /abs
      /addition
      ...

Development

Linux

Prerequisites

1. C++ compiler (gcc or clang)
2. git
3. clang-format
4. make (for Makefiles)
5. cmake (version 3.13 or higher)

Check out the library.

$ git clone https://github.com/mlcpp/Matrix.git && cd Matrix

Create build directory.

$ mkdir build

Enter into build directory.

$ cd build

Run CMake. This step will generate a Makefile in the build/ folder.

$ cmake ../

Compile all files and place the binaries inside build/examples/ folder.

$ make

Compile single file and place the binary inside build/examples/ folder.

$ make

Windows

Prerequisites

1. C++ compiler (mingw-gcc)
2. git (msysgit)
3. clang-format
4. mingw-make or mingw32-make (for Makefiles)
5. cmake (version 3.13 or higher)

Add these tools to PATH.

Note: Run the following commands in msysgit terminal.

Check out the library.

$ git clone https://github.com/mlcpp/Matrix.git && cd Matrix

Create build directory.

$ mkdir build

Enter into build directory.

$ cd build

Run CMake. This step will generate a Makefile in the build/ folder.

$ cmake ../

Compile all files and place the binaries inside build/examples/ folder.

$ mingw-make # use mingw32-make if using 32-bit Windows

Compile single file and place the binary inside build/examples/ folder.

$ mingw-make # use mingw32-make if using 32-bit Windows

Benchmarking

To compile g_benchmark binaries:

Linux: make benchmarks

Windows: mingw32-make benchmarks

Testing

To compile g_test binaries:

Linux: make tests

Windows: mingw32-make tests

Quick Start Guide

This guide contains detailed explanation of all the methods and functions available and how to use them.

More detailed information about how to use these methods and functions is available in /examples.

Note: Here d-type means any C++ data type.

Initializers

Many initializer functions are provided that return Matrix object.

Slicing

Matrix objects can be sliced like NumPy arrays.

Printing/Viewing

A number of methods are provided to print/view a Matrix object in different ways.

Indexing

Indexing can be used to get or assign value to a particular element.

For example, let there is a Matrix object named mat. If we want to get the value at index (5,3) we can do this as follows:

Similarly, if we want to assign the value to index (5,3) we can do this by:

Operators

Support for almost all standard C++ operators is provided. This includes:

• Arithmetic operators (+, -, \*, /)
• Compound assignment operatos (+=, -=, \*=, /=)
• Increment and decrement operators (++, --)
• Relational and Comparison operators (==, !=, <, >, <=, >=)
• Unary Minus (-)

Broadcasting

Broadcasting is in-built in the Basic Mathematical operations i.e., addition, subtraction, multiplication and division.

Following binary operations are possible:

• Matrix = Matrix @ Matrix
• Matrix = Matrix @ Vector
• Matrix = Matrix @ Scalar

where, @ is any operator from (+, -, *, /)

Note: Vector is a Matrix object where row length or column length is equal to 1.

Following unary operations are possible:

• Matrix = @Matrix

where, @ is any operator from (-)

Minimum, Maximum

Mathematical Operations

Note: Broadcasting in power() methods works in the same way as in Basic Mathematical operations.

Statistical Operations

Matrix Algebra

Miscellaneous