mlpy.mdp.continuous.casml.CASMLRevisionMethod

class mlpy.mdp.continuous.casml.CASMLRevisionMethod(rho=None, plot_revision=None, plot_revision_params=None)[source]

Bases: mlpy.knowledgerep.cbr.methods.IRevisionMethod

The revision method implementation for CASML.

The solutions provided by the query case is evaluated and information about whether the solution has or has not provided a desired outcome is gathered.

Parameters:

rho : float, optional

The permitted error of the similarity measure. Default is 0.99.

plot_revision : bool, optional

Whether to plot the vision step or not. Default is False.

plot_revision_params : {‘origin_to_query’, ‘original_origin’}

Parameters used for plotting. Valid parameters are:

origin_to_query

Which moves the origins of all actions to the query case’s origin.

original_origin

The origins remain unchanged and the states are plotted at their original origins.

Notes

The CASML revision method further narrows down the solutions to the query case by identifying whether the actions are similar by ensuring that the actions are cosine similar within the permitted error \rho:

d(c_{q.\text{action}}, c.\text{action}) >= \rho

Methods

execute(case, case_matches, **kwargs) Execute the revision step.
plot_data(case, case_matches, **kwargs) Plot the data during the revision step.