资 源 简 介
Introduction
This software is used for educational purposes in order to teach how to implement the various Hidden Markov Model (HMM) algorithms in C and CUDA. It contains the necessary files to run the following algorithms:
* Forward Algorithm (fo)
* Viterbi Algorithm (vit)
* Baum-Welch Algorithm (bwa)
Additionally, it contains various test files needed to test the algorithms in both C and CUDA along with timer functions so that the user can see the time difference between C and CUDA. It is not recommended to use the algorithms in a production environment as they have not been completely optimized.
Note - This code was originally based on the work done by Chuan Liu, which can be found here:
http://liuchuan.org/pub/cuHMM.pdf
http://code.google.com/p/chmm
Installation
Prerequisites:
* NVIDIA graphics card (CUDA-capable)
* Linux (I used Ubuntu 10.04 32-bit)
1) Install the CUDA SDK
* There are many guide on the Internet on
