Python module to train GMMs using CUDA (via CUDAMat)
Dependencies
Installation
Example usage
Documentation
###Dependencies
- Not Windows (only tested on Linux and Mac)
- CUDA 6.0+ (only tested with 6.0)
- numpy
- CUDAMat, avaiable here: https://github.com/cudamat/cudamat.git
- nose (optional, for running tests)
###Installation
Clone ggmm and CUDAMat in local install path:
cd ${INSTALL_PATH}
git clone https://github.com/ebattenberg/ggmm.git
git clone https://github.com/cudamat/cudamat.gitCompile and install CUDAMat:
cd ${INSTALL_PATH}/cudamat
sudo python setup.py installRun CUDAMat tests (optional, requires nose):
cd ${INSTALL_PATH}/cudamat
nosetests
Run ggmm tests (optional, requires nose):
cd ${INSTALL_PATH}/ggmm
nosetests###Example Usage
import ggmm.gpu as ggmm
X = some_module.load_training_data()
# N - training examples
# D - data dimension
# K - number of GMM components
N, D = X.shape
K = 128
ggmm.init()
gmm = ggmm.GMM(K,D)
thresh = 1e-3 # convergence threshold
n_iter = 20 # maximum number of EM iterations
init_params = 'wmc' # initialize weights, means, and covariances
# train GMM
gmm.fit(X, thresh, n_iter, init_params=init_params)
# retrieve parameters from trained GMM
weights = gmm.get_weights()
means = gmm.get_means()
covars = gmm.get_covars()
# compute posteriors of data
posteriors = gmm.compute_posteriors(X)Documentation available here