# doc-cache created by Octave 6.4.0
# name: cache
# type: cell
# rows: 3
# columns: 19
# name: <cell-element>
# type: sq_string
# elements: 1
# length: 28
mvn_bregmancentroid_geodesic


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 394
MVN_BREGMANCENTROID_GEODESIC   Computes an arbitrary weighted KL centroid on the geodesic linking between two sided centroids c1 and c2.

   [c] = MVN_BREGMANCENTROID_GEODESIC(d, m1, m2) computes the KL centroid
   for the given sided centroids m1 and m2 weighting m1 with d and m2
   with (1-d).

   See: On the centroids of symmetrized bregman divergences, Nielsen, F.
   and Nock, R., 2007.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_BREGMANCENTROID_GEODESIC   Computes an arbitrary weighted KL centroid on ...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 27
mvn_bregmancentroid_kl_left


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 256
MVN_BREGMANCENTROID_KL_LEFT    Compute the left sided Kullback-Leibler (KL) centroid given mvn models. (GRADIENT SPACE CENTER OF MASS)

   [c] = MVN_BREGMANCENTROID_KL_LEFT(models) computes the left sided KL
   centroid for the given multivariate normals.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_BREGMANCENTROID_KL_LEFT    Compute the left sided Kullback-Leibler (KL) c...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 28
mvn_bregmancentroid_kl_right


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 210
MVN_BREGMANCENTROID_RIGHT    Compute the right centroid given mvn models ("center of mass").

   [c] = MVN_BREGMANCENTROID_RIGHT(models) computes the right sided
   centroid for the given multivariate normals.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_BREGMANCENTROID_RIGHT    Compute the right centroid given mvn models ("ce...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 23
mvn_bregmancentroid_skl


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 341
MVN_BREGMANCENTROID_SKL    Compute the symmetric Kullback-Leibler (KL) centroid given mvn models.

   [c] = MVN_BREGMANCENTROID_SKL(models) computes the SKL centroid 
   for the given multivariate normals.

   [c] = MVN_BREGMANCENTROID_SKL(models, 1) computes the SKL centroid 
   mid-point approximation for the given multivariate normals.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_BREGMANCENTROID_SKL    Compute the symmetric Kullback-Leibler (KL) centro...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 10
mvn_div_bc


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 189
MVN_DIV_BC Compute the Bhattacharyya coefficient between two
   multivariate normals.

   [d] = MVN_DIV_BC(m1, m2) computes the BC between two
   multivariate normals. d is never negative.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_DIV_BC Compute the Bhattacharyya coefficient between two
   multivariate ...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 10
mvn_div_js


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 288
MVN_DIV_JS Compute the Jensen-Shannon (JS) divergence between two multivariate normals.

   [d] = MVN_DIV_JS(m1, m2) computes the JS divergence between two
   multivariate normals. d is never negative.

   The JS divergence is defined as:
       d = 0.5*KL(m1, m1+m2) + 0.5*KL(m2, m1+m2)



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_DIV_JS Compute the Jensen-Shannon (JS) divergence between two multivariat...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 10
mvn_div_kl


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 172
MVN_DIV_KL  Compute the Kullback-Leibler (KL) divergence.

   [d] = MVN_DIV_KL(m1, m2) computes the KL divergence between two
   multivariate normals. d is never negative.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 57
MVN_DIV_KL  Compute the Kullback-Leibler (KL) divergence.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 11
mvn_div_skl


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 317
MVN_DIV_SKL  Compute the symmetric Kullback-Leibler (KL) divergence.

   [d] = MVN_DIV_SKL(m1, m2) computes the SKL divergence between two
   multivariate normals. d is never negative.

   The SKL divergence is defined as:
       d = 0.5*KL(m1, m2) + 0.5*KL(m2, m1)

   Note: This function computes a faster version.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 68
MVN_DIV_SKL  Compute the symmetric Kullback-Leibler (KL) divergence.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 10
mvn_divmat


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 304
MVN_DIVMAT  Compute a distance/divergence matrix using the specified MODELS and the given divergence T.

   [d] = MVN_DIVMAT(models, t) Computes a divergence matrix using the given
   mvn MODELS models and divergence T. T can be 'kl_left', 'kl_right', 
   'skl', 'js', 'js_kl'. It defaults to the 'skl'.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_DIVMAT  Compute a distance/divergence matrix using the specified MODELS a...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 11
mvn_entropy


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 128
MVN_ENTROPY Compute the Entropy of the given multivariate normal.

   [h] = MVN_ENTROPY(m) computes the entropy of mvn model M.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 65
MVN_ENTROPY Compute the Entropy of the given multivariate normal.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 12
mvn_fn2class


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 14
MVN_FN2CLASS 



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 14
MVN_FN2CLASS 




# name: <cell-element>
# type: sq_string
# elements: 1
# length: 12
mvn_ismetric


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 324
MVN_ISMETRIC Analyzes the distance matrix D and returns the percentage of triples fulfilling the triangle inequality.

   ATTENTION: Symmetry and Identity is presumed.

   ismetric = MVN_ISMETRIC(d) Analyzes the given distance matrix and returns
   the percentage of all possible triples fulfilling the triangle inequality.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_ISMETRIC Analyzes the distance matrix D and returns the percentage of tri...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 10
mvn_kmeans


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 500
MVN_KMEANS Compute K-Means clustering on the given mvn models in respect to the given divergence. Standard is the symmetric Kullback-Leibler divergence between the models.

   [centers, assig] = MVN_KMEANS_SKL(models, k, type) computes a k-means
   clustering using the given the SKL divergence. We return k mvn
   centers and the assignment of the models to their centers.

   See: Clustering with Bregman divergences, A. Banerjee et al., The
   Journal of Machine Learning Research 2005, Volume 6.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_KMEANS Compute K-Means clustering on the given mvn models in respect to t...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 12
mvn_knnclass


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 13
MVN_KNNCLASS



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 13
MVN_KNNCLASS




# name: <cell-element>
# type: sq_string
# elements: 1
# length: 7
mvn_new


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 145
MVN_NEW  Initialize a mvn struct

   [d] = MVN_NEW(cov, m) initializes a MVN struct using the given covariance
   matrix COV and mean vector M.




# name: <cell-element>
# type: sq_string
# elements: 1
# length: 32
MVN_NEW  Initialize a mvn struct



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 11
mvn_som_skl


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 655
MVN_SOM_SKL  Computes a NxN SOM using the MVN models and the
   Symmetric Kullback-Leibler divergence. 

The algorithm was published in 2010:
   ``Islands of Gaussians: The Self Organizing Map and Gaussian Music Similarity Features'', D. Schnitzer, A. Flexer, G. Widmer and M. Gasser, Proceedings of the 11th International Society for Music Information Retrieval Conference, 2010.

   [grid D] = MVN_SOM_SKL(models, n) Select the models to compute
   the NxN SOM.

   [grid D] = MVN_SOM_SKL(models, n, global_training, local_training)
   Select the models to compute the NxN SOM; global/local_training sets the global/local training iterations of the SOM



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_SOM_SKL  Computes a NxN SOM using the MVN models and the
   Symmetric Kul...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 8
mvn_test


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 78
MVN_TEST  Load a testcollection and prepare for using with the mvn functions.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 78
MVN_TEST  Load a testcollection and prepare for using with the mvn functions.




# name: <cell-element>
# type: sq_string
# elements: 1
# length: 13
mvn_traceprod


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 206
MVN_TRACEPROD  Compute Matrix trace of the matrix product of two symmetric matrices a and b.

   [c] = MVN_TRACEPROD(A, B) computes the trace of the symmetric matices
   c = trace(A*B) in an efficient way.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 80
MVN_TRACEPROD  Compute Matrix trace of the matrix product of two symmetric ma...



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 11
mvn_version


# name: <cell-element>
# type: sq_string
# elements: 1
# length: 66
MVN_VERSION  

   [v] = MVN_VERSION() Returns the version of MVN.



# name: <cell-element>
# type: sq_string
# elements: 1
# length: 13
MVN_VERSION  





