BP神經(jīng)網(wǎng)絡(luò)算法的C語言實現(xiàn)代碼.doc

精品文檔/BP神經(jīng)網(wǎng)絡(luò)算法,c語言版本/VS2010下，無語法錯誤，可直接運行/添加了簡單注釋/歡迎學(xué)習(xí)交流#include #include #include #include #define N_Out 2 /輸出向量維數(shù)#define N_In 3/輸入向量維數(shù)#define N_Sample 6 /樣本數(shù)量/BP人工神經(jīng)網(wǎng)絡(luò)typedef struct int LayerNum;/中間層數(shù)量double vN_In50;/中間層權(quán)矩陣i,中間層節(jié)點最大數(shù)量為50double w50N_Out;/輸出層權(quán)矩陣double StudyRate;/學(xué)習(xí)率double Accuracy;/精度控制參數(shù)int MaxLoop;/最大循環(huán)次數(shù) BPNet;/Sigmoid函數(shù)double fnet(double net) return 1/(1+exp(-net);/初始化int InitBpNet(BPNet *BP);/訓(xùn)練BP網(wǎng)絡(luò)，樣本為x，理想輸出為yint TrainBpNet(BPNet *BP, double xN_SampleN_In, int yN_SampleN_Out) ;/使用BP網(wǎng)絡(luò)int UseBpNet(BPNet *BP);/主函數(shù)int main()/訓(xùn)練樣本double xN_SampleN_In = 0.8,0.5,0, 0.9,0.7,0.3,1,0.8,0.5,0,0.2,0.3,0.2,0.1,1.3,0.2,0.7,0.8; /理想輸出int yN_SampleN_Out = 0,1,0,1,0,1,1,1,1,0,1,0; BPNet BP;InitBpNet(&BP); /初始化BP網(wǎng)絡(luò)結(jié)構(gòu)TrainBpNet(&BP, x, y); /訓(xùn)練BP神經(jīng)網(wǎng)絡(luò)UseBpNet(&BP); /測試BP神經(jīng)網(wǎng)絡(luò)return 1; /使用BP網(wǎng)絡(luò)int UseBpNet(BPNet *BP) double InputN_In;double Out150; double Out2N_Out; /Out1為中間層輸出,Out2為輸出層輸出/持續(xù)執(zhí)行，除非中斷程序while (1) printf(請輸入3個數(shù)：n);int i, j;for (i = 0; i N_In; i+)scanf_s(%f, &Inputi);double Tmp;for (i = 0; i (*BP).LayerNum; i+) Tmp = 0;for (j = 0; j N_In; j+)Tmp += Inputj * (*BP).vji;Out1i = fnet(Tmp);for (i = 0; i N_Out; i+) Tmp = 0;for (j = 0; j (*BP).LayerNum; j+)Tmp += Out1j * (*BP).wji;Out2i = fnet(Tmp);printf(結(jié)果： );for (i = 0; i N_Out; i+)printf(%.3f , Out2i);printf(n);return 1;/訓(xùn)練BP網(wǎng)絡(luò)，樣本為x，理想輸出為yint TrainBpNet(BPNet *BP, double xN_SampleN_In, int yN_SampleN_Out) double f = (*BP).Accuracy; /精度控制參數(shù)double a = (*BP).StudyRate; /學(xué)習(xí)率int LayerNum = (*BP).LayerNum; /中間層節(jié)點數(shù)double vN_In50, w50N_Out; /權(quán)矩陣double ChgH50, ChgON_Out; /修改量矩陣double Out150, Out2N_Out; /中間層和輸出層輸出量int MaxLoop = (*BP).MaxLoop; /最大循環(huán)次數(shù)int i, j, k, n;double Tmp;for (i = 0; i N_In; i+)/ 復(fù)制結(jié)構(gòu)體中的權(quán)矩陣 for (j = 0; j LayerNum; j+)vij = (*BP).vij;for (i = 0; i LayerNum; i+)for (j = 0; j f & n MaxLoop; n+) e = 0;for (i= 0; i N_Sample; i+) /計算中間層輸出向量for (k= 0; k LayerNum; k+) Tmp = 0;for (j = 0; j N_In; j+)Tmp = Tmp + xij * vjk; Out1k = fnet(Tmp);/計算輸出層輸出向量for (k = 0; k N_Out; k+) Tmp = 0;for (j = 0; j LayerNum; j+)Tmp = Tmp + Out1j * wjk;Out2k = fnet(Tmp);/計算輸出層的權(quán)修改量 for (j = 0; j N_Out; j+) ChgOj = Out2j * (1 - Out2j) * (yij - Out2j);/計算輸出誤差for (j = 0; j N_Out ; j+) e = e + (yij - Out2j) * (yij - Out2j);/計算中間層權(quán)修改量for (j = 0; j LayerNum; j+) Tmp = 0;for (k = 0; k N_Out; k+)Tmp = Tmp + wjk * ChgOk;ChgHj = Tmp * Out1j * (1 - Out1j);/修改輸出層權(quán)矩陣for (j = 0; j LayerNum; j+) for (k = 0; k N_Out; k+)wjk = wjk + a * Out1j * ChgOk; for (j = 0; j N_In; j+)for (k = 0; k LayerNum; k+)vjk = vjk + a * xij * ChgHk; if (n % 10 = 0)printf(誤差 : %fn, e);printf(總共循環(huán)次數(shù)：%dn, n);printf(調(diào)整后的中間層權(quán)矩陣：n);for (i = 0; i N_In; i+) for (j = 0; j LayerNum; j+)printf(%f , vij); printf(n);printf(調(diào)整后的輸出層權(quán)矩陣：n);for (i = 0; i LayerNum; i+) for (j = 0; j N_Out; j+)printf(%f , wij); printf(n);/把結(jié)果復(fù)制回結(jié)構(gòu)體 for (i = 0; i N_In; i+) for (j = 0; j LayerNum; j+)(*BP).vij = vij;for (i = 0; i LayerNum; i+)for (j = 0; j N_Out; j+)(*BP).wij = wij;printf(BP網(wǎng)絡(luò)訓(xùn)練結(jié)束！n);return 1;/初始化int InitBpNet(BPNet *BP) printf(請輸入中間層節(jié)點數(shù)，最大數(shù)為100：n); scanf_s(%d, &(*BP).LayerNum);printf(請輸入學(xué)習(xí)率：n);scanf_s(%lf, &(*BP).StudyRate); /(*BP).StudyRate為double型數(shù)據(jù)，所以必須是lfprintf(請輸入精度控制參數(shù)：n);scanf_s(%lf, &(*BP).Accuracy);printf(請輸入最大循環(huán)次數(shù)：n);scanf_s(%d, &(*BP).MaxLoop);int i, j;srand(unsigned)time(NULL);for (i = 0; i N_In; i+) for (j = 0; j (*BP).LayerNum; j+)(*BP).vij = rand() / (double)(RA