GmmModel train¶
-
train(self, frame, observation_columns, column_scalings, k=2, max_iterations=100, convergence_tol=0.01, seed=4729768646873607665)¶ Creates a GMM Model from the train frame.
Parameters: frame : Frame
A frame to train the model on.
observation_columns : list
Columns containing the observations.
column_scalings : list
Column scalings for each of the observation columns. The scaling value is multiplied by the corresponding value in the observation column.
k : int32 (default=2)
Desired number of clusters. Default is 2.
max_iterations : int32 (default=100)
Number of iterations for which the algorithm should run. Default is 100.
convergence_tol : float64 (default=0.01)
Largest change in log-likelihood at which convergence iis considered to have occurred.
seed : int64 (default=4729768646873607665)
Random seed
Returns: : dict
- dict
Returns a dictionary the following fields
- cluster_size : dict
with the key being a string of the form ‘Cluster:Id’ storing the number of elements in cluster number ‘Id’
- gaussians : dict
Stores the ‘mu’ and ‘sigma’ corresponding to the Multivariate Gaussian (Normal) Distribution for each Gaussian
At training the ‘k’ cluster centers are computed.
Examples
See here for examples.