net aps.rar_aps

/* * File: net_glc_fun.c * * MATLAB Coder version : 2.6 * C/C++ source code generated on : 01-Aug-2016 17:27:42 */ /* Include files */ #include "rt_nonfinite.h" #include "net_glc_fun.h" /* Function Definitions */ /* * NET_GLC_FUN neural network simulation function. * * Generated by Neural Network Toolbox function genFunction, 01-Aug-2016 17:20:24. * * [y1] = net_glc_fun(x1) takes these arguments: * x = 3xQ matrix, input #1 * and returns: * y = 1xQ matrix, output #1 * where Q is the number of samples. * Arguments : double glc * double insulin * double cpep * Return Type : double */ double net_glc_fun(double glc, double insulin, double cpep) { double x1[3]; int k; static const double settings_gain[3] = { 1.25000562189975, 0.0031290148960008, 0.277009574089242 }; static const double settings_xoffset[3] = { 1.70000392892616, 12.9208038828615, 0.760028437825413 }; double b[19]; static const double a[19] = { 3.42093017210012, 2.9430386414708667, 2.3571952513947196, -3.5611770059023589, 0.79381211338056623, 0.97410912708967656, 1.2842025188274149, -3.5463918859922381, -6.6049692452812616, 1.6276624973847353, 4.5420637538775708, -8.29230163234388, 8.0930162556110243, 6.333343596386074, 3.6587434161244392, 2.4521883530802921, 2.7404733244778705, 4.3550533909909772, 4.1461418669175183 }; double b_b[19]; double d0; int i0; static const double b_a[57] = { 0.06721728705456724, -2.8187992819099987, -1.5486992042637486, 0.69321764292268362, -0.83208303979112563, 0.35257017104981864, -4.7694757538701813, 6.41312386839963, 9.3647630782610118, -1.6685644674594362, 11.20785624225157, -8.84708503237788, 8.32606457717145, 2.1714070788033628, 9.0205510972025369, 6.6306970701179884, 3.7257911444219833, 9.63149216966046, 2.8540696333113904, -3.6440659210095543, 0.32532652774011384, 2.5129258363036766, -5.37015093714612, -0.22679397401866927, 0.40163084653996967, -2.7216771625106255, -1.5207989604210006, -20.528289340450062, 5.76392526407906, 5.8405525707589749, -8.9583530549360066, 12.96435300553618, 8.817969633496741, 4.9032641835373925, -0.355014895455146, 1.6828881466217667, -2.1359061698092372, -0.14138356264410198, -3.6527437844691777, 2.0318828460311544, -0.18431043613623685, 1.0973703971636908, -1.726580454643373, -3.098395977316684, -2.5294910526653469, -8.2507005722383653, 14.34001884159173, 6.2805956326444905, 1.23954938230253, -7.8655624056693147, -0.2742871031877408, -4.1881136439579567, 0.59198691261361014, -2.4144146636043473, -2.1325563522664264, -7.3450960977362669, -2.658216589598263 }; static const double c_a[19] = { -2.7640140899660439, -2.9198052206587528, 4.9820111941264686, 2.15009151417837, -10.893591989564634, 6.3870060505267912, 1.9416126385739931, 0.69169494494283779, 0.52559912214508331, 0.43750130352239025, 6.2409668218565546, 0.953681996751488, 1.5330829626070177, 3.2395966293667766, -7.8726654434834984, 1.5558730719219735, -5.6362815601885323, 0.93303747457848873, 4.0101986664913758 }; /* ===== NEURAL NETWORK CONSTANTS ===== */ /* Input 1 */ x1[0] = glc; x1[1] = insulin; x1[2] = cpep; /* Layer 1 */ /* Layer 2 */ /* Output 1 */ /* ===== SIMULATION ======== */ /* Dimensions */ /* samples */ /* Input 1 */ /* ===== MODULE FUNCTIONS ======== */ /* Map Minimum and Maximum Input Processing Function */ for (k = 0; k < 3; k++) { x1[k] = (x1[k] - settings_xoffset[k]) * settings_gain[k] + -1.0; } /* Layer 1 */ memcpy(&b[0], &a[0], 19U * sizeof(double)); /* Sigmoid Symmetric Transfer Function */ for (k = 0; k < 19; k++) { d0 = 0.0; for (i0 = 0; i0 < 3; i0++) { d0 += b_a[k + 19 * i0] * x1[i0]; } b_b[k] = b[k] + d0; } /* Layer 2 */ /* Output 1 */ /* Map Minimum and Maximum Output Reverse-Processing Function */ d0 = 0.0; for (k = 0; k < 19; k++) { d0 += c_a[k] * (2.0 / (1.0 + exp(-2.0 * b_b[k])) - 1.0); } return ((1.5307560560365561 + d0) - -1.0) / 0.763367047638489 + 3.38000010222326; } /* * File trailer for net_glc_fun.c * * [EOF] */