About MatConvNet

MatConvNet was born in the Oxford Visual Geometry Group as an educational and research platform for fast prototyping of Convolutional Neural Networks.

MatConvNet is a flexible framework capable of learning large neural networks on one or multiple GPUs. Except for low-level building blocks, it is all written in MATLAB, which allows easy hacking, including creating new autodiff methods. Many research papers have used MatConvNet.

Changes

Contributors

MatConvNet is developed by several hands:

MatConvNet quality also depends on the many people using the toolbox and providing us with feedback and bug reports.

Copyright

This package was originally created by Andrea Vedaldi and Karel Lenc and it is currently developed by a small community of contributors. It is distributed under the permissive BSD license (see also the file COPYING):

Copyright (c) 2014-16 The MatConvNet team.
All rights reserved.

Redistribution and use in source and binary forms are permitted
provided that the above copyright notice and this paragraph are
duplicated in all such forms and that any documentation,
advertising materials, and other materials related to such
distribution and use acknowledge that the software was developed
by the <organization>. The name of the <organization> may not be
used to endorse or promote products derived from this software
without specific prior written permission.  THIS SOFTWARE IS
PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

Acknowledgments

The implementation of the computational blocks in this library, and in particular of the convolution operators, is inspired by Caffe.

We gratefully acknowledge the support of NVIDIA Corporation with the donation of the GPUs used to develop this software.