7 Commits

Author	SHA1	Message	Date
jlizier	42f8d01e26	Adding property of number of dimensions for the AIS multivariate calculator (and implemented in the via MI common class). Needed this to make multivariate TE embedding a bit neater, and it should be there anyway.	2018-10-18 00:51:37 +11:00
jlizier	e1bad39e4e	Added proper full definitions for the methods of the AIS multivariate interface; much of this is achieved by extending the univariate AIS interface (which mandates that the univariate methods must be supported). No current child classes required any changes as they all supported these already. Removed superfluous documentation in the AIS multivariate via MI class (as well as fixing whitespace inconsistencies therein)	2018-10-17 23:37:43 +11:00
jlizier	ca909dc60f	Adding check of dimensionality of observations supplied to multivariate AIS calculators when univariate is expected.	2018-06-19 23:45:00 +10:00
jlizier	dfe7ef0334	Altering AIS Multivariate Via Mutual Info calculator -- and child classes, now including kernel estimator -- to support auto-embedding primarily via the parent methods of the univariate AIS calculator. ... Also includes patching functionality when dimension == 1 to always use the underlying AIS univariate calculator consistently, and adding missing methods on adding observations with validity array. Also includes some minor alterations to error messages. Gaussian and Kernel Multivariate AIS calculators have the computeAdditionalBiasToRemove() method implemented now to allow this; Gaussian calculator also adds the analytic computeSignificance() method, whilst Kernel calculator also adds a property for the number of surrogates to use in bias correction.	2018-05-16 22:42:32 +10:00
jlizier	3e64fd28bd	Altering various continuous calculators to implement the EmpiricalNullDistributionComputer interface (where they're already implementing computeSignificance() etc.). Also altered the EmpiricalNullDistributionComputer methods to throw Exceptions, since the continuous calculators generally do this (and doesn't harm the discrete ones). Also involved implementing the methods in ConditionalMIMultiVariateCommon by selecting to permute the first variable by default.	2017-08-18 15:31:27 +10:00
Pedro Martinez Mediano	6f87aed980	Added preliminary version of Ragwitz optimization in multivariate AIS calculator.	2016-04-19 18:22:02 +01:00
Pedro Martinez Mediano	3da629a817	Preliminary commit of multivariate AIS calculators and unit tests. Fully functional, but not tested enough.	2016-03-15 22:36:12 +00:00