gmmp
Functions
-
int main(int argc, char *argv[])
gmmp [ option ] gmmfile [ infile ]
-l int
length of vector \((1 \le L)\)
-m int
order of vector \((0 \le L - 1)\)
-k int
number of mixtures \((1 \le K)\)
-f
use full or block covariance instead of diagonal one
gmmfile str
double-type GMM parameters
infile str
double-type input data sequencea
stdout
double-type log-probability
The input of this command is
\[ \begin{array}{cccc} \boldsymbol{x}(0), & \boldsymbol{x}(1), & \ldots, & \boldsymbol{x}(T-1), \end{array} \]where \(\boldsymbol{x}(t)\) is \(L\)-dimensional vector. The output is a sequence of log-probabilities of the input vectors:\[ \begin{array}{ccc} \log p(\boldsymbol{x}(0)), & \ldots, & \log p(\boldsymbol{x}(T-1)), \end{array} \]where\[ p(\boldsymbol{x}(t)) = \sum_{k=0}^{K-1} w_k \mathcal{N}(\boldsymbol{x}(t) \, | \, \boldsymbol{\mu}_k, \boldsymbol{\varSigma}_k), \]where \(w_k\), \(\boldsymbol{\mu}_k\), and \(\boldsymbol{\varSigma}_k\), are the parameters of GMM.In the following example, the log-probabilities of input data read from
data.dbased on 4-mixture GMM are calculetaed and then averaged.gmmp -k 4 data.gmm < data.d > data.p vstat -o 1 data.p > data.p.avg
- Parameters:
argc – [in] Number of arguments.
argv – [in] Argument vector.
- Returns:
0 on success, 1 on failure.