README.md

This is a C library which can be used in deep learning applications. It allows multiple layers to be trained and also includes the dropouts technique to avoid overfitting the data.

Installation

On Debian based systems:

sudo apt-get install gnuplot libpng-dev doxygen

On Fedora based systems:

sudo yum install gnuplot libpng-devel doxygen

To build from source:

make
sudo make install

This creates the library and installs it into /usr/local

Unit Tests

You can run the unit tests to check that the system is working as expected:

make tests
./tests

Or to check for any memory leaks:

valgrind --leak-check=full ./tests

Source Documentation

To generate source code documentation make sure that you have doxygen installed and then run the generatedocs.sh script. A subdirectory called docs will be created within which html and latex formated documentation can be found. For general usage information you can also see the manpage.

man libdeep

Examples

There are also some example programs within the examples directory. Reading the examples is the best way to learn how to use this library within your own code.

Portability

Although this software was primarily written to run on Linux-based systems it's pretty much just vanilla C code and so it should be easily portable to other platforms, such as Microsoft Windows and Mac systems. The independent random number generator should mean that results are consistent across different compilers and platforms.

Packaging

To build packages for Debian (deb) or Fedora (rpm) based systems there are two scripts, debian.sh and rpm.sh which can be used to generate packages for convenient installation.