< Master index Index for SFMI >

Index for SFMI

Functions:

 AboutABOUT M-file for About.fig
 AdjustSideiter - the number of the current iteration
 AlphaCutDetermine the lower and upper endpoints of an alpha cut of a fuzzy cluster.
 ApproxInputTrapezoidsApproximate the membership functions by trapezoids forming a Ruspini-like partition
 ApproxOutputTrapezoidsApproximate the membership functions by trapezoids forming a Ruspini-like partition
 ApproxTrapezoid
 CNFUnionMerge the sets of the antecedent parts of two rules.
 Calc1Dir
 Cluster1DPerform 1D FCM Clustering
 ClusteringPreferencesCLUSTERINGPREFERENCES M-file for ClusteringPreferences.fig
 CorrectMfSupports4RBE_DSSThe support of a set cannot overlap the core of another set.
 CorrectOverlappingMfsThe cores of two sets cannot overlap each other.
 CorrectRulebaseReplace the references to the antecedent set 'Original' with the references to the antecedent set 'New'.
 CorrectSetCoreNumber of input dimensions
 CorrectSetShapesCorrect set shapes after the introduction of the last set in order to fulfil the conditions of tuning method RBE-DSS.
 CreateExample% Create example data for fis trainig
 CreateSystemWith2RulesCreate a raw fuzzy system containing only two rules.
 CsereBerePerm-be egy n-ed rendű permutációt, Sign-be véletlen előjelet ad:
 DefaultClusteringOptionsoptions - column vector containing parameters of the FCM clustering
 DefaultTuningOptionsCreate a vector containing the default values for tuning preferences
 DefaultTwoRulesOptionsCreate a column vector containing the default values for the options
 DefineNewRule4RBE_DSSDefine a new rule in that point where the system gives the worst
 DefineNewRule4RBE_SIDefine a new rule in that point where the system gives the worst
 DefineSetsDefine the reference points of the sets and label the sets.
 DetermineAntecedentDetermine a submatrix of the matrix describing the antecedent parts of the rules.
 DisplayResults
 DoTuning
 DrawReferencePoint
 ElimineDuplicateRulesEliminate rules that are already present in the rulebase.
 Exp01Define the data for the first experiment
 ExtendRange4VEExtend the range of each linguistic variable in order to can
 ExtendSetsOfFis
 ExtendSetsOfPartition
 FcmClustering1D1D fuzzy c-means clustering
 Felfedezo---------------------------------------------------------------------
 Fis2SideCopy
 Fis2VectorTransform the structure-based description of the parameters into a row vector based one.
 GetClusterMembershipValuesv - column vector containig the cluster centers
 GetMinRangeDetermine the minum of the range of the input and output dimensions
 GetRawFISNameGETRAWFISNAME M-file for GetRawFISName.fig
 GetRuleDetermine the rules for the current output fuzzy set
 GetTuningPreferencesDetermine the tuning options using a dialog box.
 InferenceSpecificCorrection
 InitializeMembershipMatrix1DInitialize the membership matrix for 1D fuzzy c-means clustering
 InputClustering1DPerform Fuzzy C-Mean Clustering in each indata dimension separately
 IsInsideDetermines whether the reference point of the set SetNo is inside the
 JoFvez egy eléggé hasraütészserűen definiált jóságfüggvény:
 MSECalcLoad the file containing the output train data.
 MergeClustersMerge 1D fuzzy clusters)
 MergeIdenticalClustersMerge antecedent clusters whose centers are identical
 MergeInputPartitionsGenereate a common input partition in each input dimension by merging the
 MergePartRulebasesGenereate a common rule base by merging the
 MergeRulesMerge the similar rules - optimize the rulebase
 MergeRulesWithIdenticalAntecedents
 MergeSimilarSetsMerge fuzzy sets whose params does not differ more than
 MergeSimilarSetsInaPartitionMerge sets whose params does not differ more than
 MergingIndexCalculate the merging index for a given vm data point
 ObjectiveFunction
 OutputClustering1DPerform Fuzzy C-Mean Clustering in each output dimension separately (1D clusterings)
 OutputDimensionSelectOUTPUTDIMENSIONSELECT M-file for OutputDimensionSelect.fig
 ParamNoDetermine the number of parameters.
 PartlyOverlappingAntecedentsIf the antecedents are at least partly overlapping
 PlotPI
 PlotPartitionDisplay the clusters in the current dimension
 RBE_DSSFuzzy system tuning with RBE-DSS
 RBE_SIFuzzy system tuning by a varianat of the gradiant descent method and
 ReadTestInputLoads the data file containing the test data
 ReadTestOutputLoads the data file containing the test data
 RefPoints2RangeReference points should be inside the range of the linguistic variable.
 RemoveEmptyMfsRemove the empty membership functions, i.e. that elements, which
 RoundFisData
 RuleDistCalculate the ditance between two rules in each input dimension. The unit
 RuleExtractLoad the file containing the output data (train+check).
 RuleMakerRULEMAKER M-file for RuleMaker.fig
 RuspiniCorrectionCorercts the shape of the sets if the Ruspini condition is not met
 SFMISFMI M-file for SFMI.fig
 SaveData
 SeekxApproximate the abscissa of the point where a horizontal line situated at CutLevel intersects the cluster membership function
 ShowClusteringPreferencesShow the options of the clustering in the editbox in the main window
 ShowResultsCheckStopPauseShow part or whole of the results depending on the options set and check the Stop Pause state.
 ShowTuningPreferencesShow the options of the system tuning in the editbox in the main window
 ShowTwoRulesPreferencesShow the options (preferences) applied by the fuzzy system generation based on extreme
 Side2FisCopy back the new values into the fis structure
 Sigmoidx=1:1:100;
 SimplexFuzzy system tuning with the Simplex method
 SortMfsSort the membership functions in each dimension in ascending order.
 SwapMfsSwap two fuzzy sets and correct the rule base.
 SystemSelectSYSTEMSELECT M-file for SystemSelect.fig
 SystemTuning1Fuzzy system tuning with ACP + breakpoints as parameters
 SystemTuning2Fuzzy system tuning with ACP + relative distances as parameters
 SystemTuning3Fuzzy system tuning with ACP + Conserving Ruspini partition
 SzProba
 Szimplex
 Tune1Param
 TuneInferenceParametersTune inference parameters using a gradient descent approach
 TuningPreferencesACPTUNINGPREFERENCESACP M-file for TuningPreferencesACP.fig
 TuningPreferencesRBE_DSSTUNINGPREFERENCESRBE_DSS M-file for TuningPreferencesRBE_DSS.fig
 TuningPreferencesRBE_SITUNINGPREFERENCESRBE_SI M-file for TuningPreferencesRBE_SI.fig
 TuningPreferencesSimplexTUNINGPREFERENCESSIMPLEX M-file for TuningPreferencesSimplex.fig
 TuningSpecificCorrectionsMake tuning specific corrections of the parameters.
 TwoRulesPreferencesTWORULESPREFERENCES M-file for TwoRulesPreferences.fig
 UnifyMfsUnify sets with reference points closer than the identity
 ValidateCorrectParameters4TuningRBE_DSSApply validity conditions and do corrections if they
 Vector2FisTransform the row vector based description of the parameters into a structure based one.
 VerifyCorrectMfParameters
 VerifyCorrectSetParameters
 VerifyCorrectSetParameters1Defines the weakest constraints:
 VerifyCorrectSetParameters2Defines the medium constraints:
 VerifyParametersVerify the parameters of a fis structure.
 Visualize
 WriteObsObs--name
 params2tpConverts the FIS parameters to the tuned parameters vector
 pvect2variableTransform the parameters of a linguistic variable containing trapezoid
 tp2paramsConverts the the tuned parameters vector to FIS parameters
 variable2pvectTransform the parameters of a linguistic variable containing trapezoid

Dependency Graph


Generated on Thu 26-Mar-2009 13:11:45 by m2html © 2005