46 MLPTrain * mlp_init(int *topo, int nbLayers, float *inputs, float *outputs, int nbSamples)  in mlp_init()  argument
52     net->topo = malloc(nbLayers*sizeof(net->topo[0]));  in mlp_init()
54         net->topo[i] = topo[i];  in mlp_init()
55     inDim = topo[0];  in mlp_init()
56     outDim = topo[nbLayers-1];  in mlp_init()
62         net->weights[i] = malloc((topo[i]+1)*topo[i+1]*sizeof(net->weights[0][0]));  in mlp_init()
63         net->best_weights[i] = malloc((topo[i]+1)*topo[i+1]*sizeof(net->weights[0][0]));  in mlp_init()
82         for (k=0;k<topo[1];k++)  in mlp_init()
83             net->weights[0][k*(topo[0]+1)+j+1] = randn(std);  in mlp_init()
86     for (j=0;j<topo[1];j++)  in mlp_init()
90             sum += inMean[k]*net->weights[0][j*(topo[0]+1)+k+1];  in mlp_init()
91         net->weights[0][j*(topo[0]+1)] = -sum;  in mlp_init()
100         std = 1/sqrt(topo[nbLayers-2]);  in mlp_init()
101         net->weights[nbLayers-2][j*(topo[nbLayers-2]+1)] = mean;  in mlp_init()
102         for (k=0;k<topo[nbLayers-2];k++)  in mlp_init()
103             net->weights[nbLayers-2][j*(topo[nbLayers-2]+1)+k+1] = randn(std);  in mlp_init()
116     int *topo;  in compute_gradient()  local
124     topo = net->topo;  in compute_gradient()
125     inDim = net->topo[0];  in compute_gradient()
126     hiddenDim = net->topo[1];  in compute_gradient()
127     outDim = net->topo[2];  in compute_gradient()
128     W0_size = (topo[0]+1)*topo[1];  in compute_gradient()
129     W1_size = (topo[1]+1)*topo[2];  in compute_gradient()
206     int *topo = arg->net->topo;  in gradient_thread_process()  local
207     W0_size = (topo[0]+1)*topo[1];  in gradient_thread_process()
208     W1_size = (topo[1]+1)*topo[2];  in gradient_thread_process()
231     int *topo;  in mlp_train_backprop()  local
240     topo = net->topo;  in mlp_train_backprop()
241     W0_size = (topo[0]+1)*topo[1];  in mlp_train_backprop()
242     W1_size = (topo[1]+1)*topo[2];  in mlp_train_backprop()
249     topo = net->topo;  in mlp_train_backprop()
250     inDim = net->topo[0];  in mlp_train_backprop()
251     hiddenDim = net->topo[1];  in mlp_train_backprop()
252     outDim = net->topo[2];  in mlp_train_backprop()
399         mean_rate /= (topo[0]+1)*topo[1] + (topo[1]+1)*topo[2];  in mlp_train_backprop()
471     int topo[3] = {nbInputs, nbHidden, nbOutputs};  in main()  local
475     net = mlp_init(topo, 3, inputs, outputs, nbSamples);  in main()
479     printf ("static const float weights[%d] = {\n", (topo[0]+1)*topo[1] + (topo[1]+1)*topo[2]);  in main()
481     for (i=0;i<(topo[0]+1)*topo[1];i++)  in main()
488     for (i=0;i<(topo[1]+1)*topo[2];i++)  in main()
495     printf ("static const int topo[3] = {%d, %d, %d};\n\n", topo[0], topo[1], topo[2]);  in main()