1 | #include "neuroimpl-fuzzy.h" |
---|
2 | #include "neuroimpl-fuzzy-f0.h" |
---|
3 | |
---|
4 | int NI_FuzzyNeuro::countOuts(const Model *m, const Neuro *fuzzy) |
---|
5 | { |
---|
6 | int outputs=0; |
---|
7 | for(int i=0;i<m->getNeuroCount();i++) |
---|
8 | for(int in=0;in<m->getNeuro(i)->getInputCount();in++) |
---|
9 | if (m->getNeuro(i)->getInput(in)==fuzzy) outputs++; |
---|
10 | return outputs; |
---|
11 | } |
---|
12 | |
---|
13 | int NI_FuzzyNeuro::lateinit() |
---|
14 | { |
---|
15 | int i, maxOutputNr; |
---|
16 | |
---|
17 | //check correctness of given parameters: string must not be null, sets&rules number > 0 |
---|
18 | if((fuzzySetsNr<1)||(rulesNr<1)||(fuzzySetString.len()==0)||(fuzzyRulesString.len()==0)) |
---|
19 | return 0; //error |
---|
20 | |
---|
21 | // this part contains transformation of fuzzy sets |
---|
22 | fuzzySets = new double[4*fuzzySetsNr]; //because every fuzzy set consist of 4 numbers |
---|
23 | // converts fuzzy string from f0 to table of fuzzy numbers type 'double' |
---|
24 | // (fill created space with numbers taken from string) |
---|
25 | // also checks whether number of fuzzy sets in the string equals declared in the definition |
---|
26 | if (FuzzyF0String::convertStrToSets(fuzzySetString, fuzzySets, fuzzySetsNr) != 0) |
---|
27 | return 0; //error |
---|
28 | |
---|
29 | // this part contains transformation of fuzzy rules and defuzzyfication parameters |
---|
30 | rulesDef = new int[2*rulesNr]; //for each rule remembers number of inputs and outputs |
---|
31 | //check correctness of string and fill in the rulesDef |
---|
32 | if (FuzzyF0String::countInputsOutputs(fuzzyRulesString, rulesDef, rulesNr) == 0) |
---|
33 | { |
---|
34 | defuzzParam = new double[rulesNr]; // parameters used in defuzyfication process |
---|
35 | // create space for rules according to rulesDef |
---|
36 | rules = new int*[rulesNr]; //list of rules... |
---|
37 | for (i=0; i<rulesNr; i++) //...that contains rules body |
---|
38 | { |
---|
39 | rules[i] = new int[2*(rulesDef[2*i]+rulesDef[2*i+1])]; //each rule can have different number of inputs and outputs |
---|
40 | defuzzParam[i] = 0; //should be done a little bit earlier, but why do not use this loop? |
---|
41 | } |
---|
42 | // fill created space with numbers taken from string |
---|
43 | if (FuzzyF0String::convertStrToRules(fuzzyRulesString, rulesDef, rules, fuzzySetsNr, rulesNr, maxOutputNr) != 0) |
---|
44 | return 0; //error |
---|
45 | } |
---|
46 | else |
---|
47 | return 0; //error |
---|
48 | |
---|
49 | setChannelCount(countOuts(neuro->owner, neuro)); |
---|
50 | return 1; //success |
---|
51 | } |
---|
52 | |
---|
53 | NI_FuzzyNeuro::~NI_FuzzyNeuro() |
---|
54 | { |
---|
55 | if(rules) //delete rows and columns of **rules |
---|
56 | { |
---|
57 | for (int i=0; i<rulesNr; i++) SAFEDELETEARRAY(rules[i]) |
---|
58 | SAFEDELETEARRAY(rules) |
---|
59 | } |
---|
60 | SAFEDELETEARRAY(defuzzParam) |
---|
61 | SAFEDELETEARRAY(rulesDef) |
---|
62 | SAFEDELETEARRAY(fuzzySets) |
---|
63 | } |
---|
64 | |
---|
65 | int NI_FuzzyNeuro::GetFuzzySetParam(int set_nr, double &left, double &midleft, double &midright, double &right) |
---|
66 | { |
---|
67 | if ( (set_nr>=0) && (set_nr<fuzzySetsNr) ) |
---|
68 | { |
---|
69 | left = fuzzySets[4*set_nr]; |
---|
70 | midleft = fuzzySets[4*set_nr+1]; |
---|
71 | midright = fuzzySets[4*set_nr+2]; |
---|
72 | right = fuzzySets[4*set_nr+3]; |
---|
73 | return 0; |
---|
74 | } |
---|
75 | else |
---|
76 | return 1; |
---|
77 | }; |
---|
78 | |
---|
79 | /** Function conduct fuzzyfication of inputs and calculates - according to rules - crisp multi-channel output */ |
---|
80 | void NI_FuzzyNeuro::go() |
---|
81 | { |
---|
82 | if (Fuzzyfication()!=0) |
---|
83 | return; |
---|
84 | if (Defuzzyfication()!=0) |
---|
85 | return; |
---|
86 | }; |
---|
87 | |
---|
88 | /** |
---|
89 | * Function conduct fuzzyfication process - calculates minimum membership function (of every input) for each rule, |
---|
90 | * and writes results into defuzzParam - variable that contains data necessary for defuzzyfication |
---|
91 | */ |
---|
92 | int NI_FuzzyNeuro::Fuzzyfication() |
---|
93 | { |
---|
94 | int i, j, nrIn, inputNr, nrFuzzySet; |
---|
95 | double minimumCut; // actual minimal level of cut (= min. membership function) |
---|
96 | |
---|
97 | // sets defuzzyfication parameters for each rule: |
---|
98 | for (i=0; i<rulesNr; i++) |
---|
99 | { |
---|
100 | nrIn = rulesDef[2*i]; // nr of inputs in rule #i |
---|
101 | minimumCut = 2; // the highest value of membership function is 1.0, so this value will definitely change |
---|
102 | for (j=0; (j<nrIn)&&(minimumCut>0); j++) //minimumCut can not be <0, so if =0 then stop calculations |
---|
103 | { |
---|
104 | nrFuzzySet = rules[i][j*2 + 1]; // j*2 moves pointer through each output, +1 moves to nr of fuzzy set |
---|
105 | inputNr = rules[i][j*2]; // as above but gives input number |
---|
106 | minimumCut = min( minimumCut, TrapeziumFuzz(nrFuzzySet, getWeightedInputState(inputNr))); // value of membership function for this input and given fuzzy set |
---|
107 | } |
---|
108 | if ( (minimumCut>1) || (minimumCut<0) ) |
---|
109 | return 1; |
---|
110 | defuzzParam[i] = minimumCut; |
---|
111 | } |
---|
112 | return 0; |
---|
113 | }; |
---|
114 | |
---|
115 | /** |
---|
116 | * Function calculates value of the membership function of the set given by wchich_fuzzy_set for given crisp value input_val |
---|
117 | * In other words, this function fuzzyficates given crisp value with given fuzzy set, returning it's membership function |
---|
118 | * @param which_fuzzy_set - 0 < number of set < fuzzySetsNr |
---|
119 | * @param input_val - crisp value of input in range <-1; 1> |
---|
120 | * @return value of membership function (of given input for given set) in range <0;1> or, if error occur, negative value |
---|
121 | */ |
---|
122 | double NI_FuzzyNeuro::TrapeziumFuzz(int which_fuzzy_set, double input_val) |
---|
123 | { |
---|
124 | double range=0, left=0, midleft=0, midright=0, right=0; |
---|
125 | |
---|
126 | if ( (which_fuzzy_set < 0) || (which_fuzzy_set > fuzzySetsNr) ) |
---|
127 | return -2; |
---|
128 | if ( (input_val < -1) || (input_val > 1) ) |
---|
129 | return -3; |
---|
130 | |
---|
131 | if (GetFuzzySetParam(which_fuzzy_set, left, midleft, midright, right) != 0) |
---|
132 | return -4; |
---|
133 | |
---|
134 | if ( (input_val < left) || (input_val > right) ) // greather than right value |
---|
135 | return 0; |
---|
136 | else if ( (input_val >= midleft) && (input_val <= midright) ) // in the core of fuzzy set |
---|
137 | return 1; |
---|
138 | else if ( (input_val >= left) && (input_val < midleft) ) // at the left side of trapezium |
---|
139 | { |
---|
140 | range = fabs(midleft - left); |
---|
141 | return fabs(input_val-left)/((range>0)?range:1); // quotient of distance between input and extreme left point of trapezium and range of rising side, or 1 |
---|
142 | } |
---|
143 | else if ( (input_val > midright) && (input_val <= right) ) // at the right side of trapezium |
---|
144 | { |
---|
145 | range = fabs(right - midright); |
---|
146 | return fabs(right-input_val)/((range>0)?range:1); // quotient of distance between input and extreme right point of trapezium and range of falling side, or 1 |
---|
147 | }; |
---|
148 | |
---|
149 | // should not occur |
---|
150 | return 0; |
---|
151 | |
---|
152 | }; |
---|
153 | |
---|
154 | /** |
---|
155 | * Function conducts defuzzyfication process: multi-channel output values are calculates with singleton method (method of high). |
---|
156 | * For each rules, all outputs fuzzy sets are taken and cut at 'cut-level', that is at minumum membership function (of current rule). |
---|
157 | * For all neuro pseudo-outputs, answer is calculated according to prior computations. |
---|
158 | * In fact, there is one output with multi-channel answer and appropriate values are given to right channels. |
---|
159 | */ |
---|
160 | int NI_FuzzyNeuro::Defuzzyfication() |
---|
161 | { |
---|
162 | int i, j, nrIn, nrOut, out, set, outputsNr; |
---|
163 | double *numerators, *denominators, midleft, midright, unimp; |
---|
164 | |
---|
165 | outputsNr = getChannelCount(); |
---|
166 | |
---|
167 | numerators = new double[outputsNr]; |
---|
168 | denominators = new double[outputsNr]; |
---|
169 | |
---|
170 | for(i=0;i<outputsNr;i++) numerators[i] = denominators[i] = 0; |
---|
171 | |
---|
172 | // for each rule... |
---|
173 | for (i=0; i<rulesNr; i++) |
---|
174 | { |
---|
175 | nrIn = rulesDef[2*i]; // number of inputs in rule #i |
---|
176 | nrOut = rulesDef[2*i + 1]; // number of outputs in rule #i |
---|
177 | // ...calculate each output's product of middle fuzzy set value and minimum membership function (numerator) and sum of minimum membership function (denominator) |
---|
178 | for (j=0; j<nrOut; j++) |
---|
179 | { |
---|
180 | out = rules[i][2*nrIn + 2*j]; //number of j-output |
---|
181 | set = rules[i][2*nrIn + 2*j + 1]; //number of fuzzy set attributed to j-output |
---|
182 | if (GetFuzzySetParam(set, unimp, midleft, midright, unimp) != 0) // gets range of core of given fuzzy set |
---|
183 | { SAFEDELETEARRAY(denominators) SAFEDELETEARRAY(numerators) return 1; } |
---|
184 | //defuzzParam[i] = minimum membership function for rule #i - calculated in fuzzyfication block |
---|
185 | // defuzzyfication method of singletons (high): (fuzzy set modal value) * (minimum membership value) |
---|
186 | numerators[out] += ((midleft + midright)/2.0) * defuzzParam[i]; |
---|
187 | denominators[out] += defuzzParam[i]; |
---|
188 | } |
---|
189 | } |
---|
190 | |
---|
191 | for (i=0; i<outputsNr; i++) |
---|
192 | { |
---|
193 | if (denominators[i] == 0) |
---|
194 | setState(0, i); |
---|
195 | else |
---|
196 | setState(numerators[i]/denominators[i], i); |
---|
197 | } |
---|
198 | |
---|
199 | SAFEDELETEARRAY(denominators) |
---|
200 | SAFEDELETEARRAY(numerators) |
---|
201 | |
---|
202 | return 0; |
---|
203 | }; |
---|
204 | |
---|