dlearn ~dev

Dlearn

High-level Linear algebra and scientific computing package in D.

Table of contents

• State of the Project
• Getting Started
• Prerequisites
• Using with dub
• Usage
• Testing
• Building the Documentation
• Contributing
• License

State of the Project

The API is in very early stages and is subject to change without notice. If you are using it in your own projects, please pin to an exact version.

Getting Started

Prerequisites

1. dmd / ldc, a D compiler
2. dub, D's package manager
3. Link with CBLAS / LAPACK

Using with dub

To use this package, run the following command in the root directory of your project

dub add dlearn

Or manually add the following to your dub.json

"dlearn": "~>0.0.2"

See dub for details.

Single File

To use dlearn as script, use dub run --single example.d, where the contents of example.d are as follows:

#!/usr/bin/env dub
/+ dub.json:
{
    "dependencies": {"dlearn": "*"}
}
+/

void main()
{
    import std : writeln;
    import dlearn;

    dlearn.arrange(5).writeln;
}

Usage

import std : writeln;
import dlearn.allocation : ones, eye;
import dlearn.math : sinh;
import dlearn.linalg : matrixMultiply;

auto x = ones!double(2, 2);
auto y = x.matrixMultiply(eye!double(2));
y.sinh.writeln;

For more examples, please refer to the Documentation.

Testing

To build the package and run all unit tests (in parallel if possible):

dub test -q --parallel

Building the Documentation

Dlearn uses ddox documentation generator which you can build with:

dub build -b ddox

Or you can build and then run a webserver with:

dub run -b ddox

Contributing

If you find a bug, please submit an issue.

Any and all contributions are appreciated. If you think of a feature you'd like added, or how we can improve the project, submit an issue too.

License

dlearn is distributed under the MIT license, as found in LICENSE.