osh / gr-theano

Rapid GPU Accelerated Blocks for GNU Radio

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

GNU Radio Theano block library!

This OOT module contains a number of GNU Radio blocks which 
leverage the theano library to accelerate signal processing 
code on graphcis cards typically using Cuda as a backend.

Enjoy!
Tim O'Shea <tim.oshea753@gmail.com>

Current Blocks:
~~~~~~~~~~~~
Signal Source (sinusoid complex64)
Fading Model (single tap complex64)
FIR Filter (float32)
FFT (complex64)

Information:
~~~~~~~~~~~~
For the latest verison please see 
https://github.com/osh/gr-theano

For more information on the theano project please see 
https://github.com/Theano/Theano
http://deeplearning.net/software/theano/
http://nbviewer.ipython.org/github/craffel/theano-tutorial/blob/master/Theano%20Tutorial.ipynb
https://archive.org/details/Scipy2010-JamesBergstra-TransparentGpuComputingWithTheano

Howto build:
~~~~~~~~~~~~
1. Install GNU Radio 3.7+
2. Install Theano
    pip install --upgrade --no-deps git+git://github.com/Theano/Theano.git
3. Install gr-threano as with any GNU Radio OOT Module




About

Rapid GPU Accelerated Blocks for GNU Radio

License:GNU General Public License v3.0


Languages

Language:CMake 69.0%Language:Python 25.4%Language:C++ 4.4%Language:C 1.2%