Fixed some problems with visualizing networks

Joseph Redmon

2015-07-24 fdcc096694791f92d2cff0e4fd74fbabfca12936

 src/network.c

@@ -10,7 +10,9 @@
#include "convolutional_layer.h"
#include "deconvolutional_layer.h"
#include "detection_layer.h"
#include "normalization_layer.h"
#include "maxpool_layer.h"
#include "avgpool_layer.h"
#include "cost_layer.h"
#include "softmax_layer.h"
#include "dropout_layer.h"
@@ -27,6 +29,8 @@
            return "connected";
        case MAXPOOL:
            return "maxpool";
        case AVGPOOL:
            return "avgpool";
        case SOFTMAX:
            return "softmax";
        case DETECTION:
@@ -39,6 +43,8 @@
            return "cost";
        case ROUTE:
            return "route";
        case NORMALIZATION:
            return "normalization";
        default:
            break;
    }
@@ -62,10 +68,15 @@
    int i;
    for(i = 0; i < net.n; ++i){
        layer l = net.layers[i];
        if(l.delta){
            scal_cpu(l.outputs * l.batch, 0, l.delta, 1);
        }
        if(l.type == CONVOLUTIONAL){
            forward_convolutional_layer(l, state);
        } else if(l.type == DECONVOLUTIONAL){
            forward_deconvolutional_layer(l, state);
        } else if(l.type == NORMALIZATION){
            forward_normalization_layer(l, state);
        } else if(l.type == DETECTION){
            forward_detection_layer(l, state);
        } else if(l.type == CONNECTED){
@@ -78,6 +89,8 @@
            forward_softmax_layer(l, state);
        } else if(l.type == MAXPOOL){
            forward_maxpool_layer(l, state);
        } else if(l.type == AVGPOOL){
            forward_avgpool_layer(l, state);
        } else if(l.type == DROPOUT){
            forward_dropout_layer(l, state);
        } else if(l.type == ROUTE){
@@ -112,13 +125,20 @@

float get_network_cost(network net)
{
    if(net.layers[net.n-1].type == COST){
        return net.layers[net.n-1].output[0];
    int i;
    float sum = 0;
    int count = 0;
    for(i = 0; i < net.n; ++i){
        if(net.layers[net.n-1].type == COST){
            sum += net.layers[net.n-1].output[0];
            ++count;
        }
        if(net.layers[net.n-1].type == DETECTION){
            sum += net.layers[net.n-1].cost[0];
            ++count;
        }
    }
    if(net.layers[net.n-1].type == DETECTION){
        return net.layers[net.n-1].cost[0];
    }
    return 0;
    return sum/count;
}

int get_predicted_class_network(network net)
@@ -147,8 +167,12 @@
            backward_convolutional_layer(l, state);
        } else if(l.type == DECONVOLUTIONAL){
            backward_deconvolutional_layer(l, state);
        } else if(l.type == NORMALIZATION){
            backward_normalization_layer(l, state);
        } else if(l.type == MAXPOOL){
            if(i != 0) backward_maxpool_layer(l, state);
        } else if(l.type == AVGPOOL){
            backward_avgpool_layer(l, state);
        } else if(l.type == DROPOUT){
            backward_dropout_layer(l, state);
        } else if(l.type == DETECTION){
@@ -167,9 +191,9 @@

float train_network_datum(network net, float *x, float *y)
{
    #ifdef GPU
#ifdef GPU
    if(gpu_index >= 0) return train_network_datum_gpu(net, x, y);
    #endif
#endif
    network_state state;
    state.input = x;
    state.delta = 0;
@@ -266,6 +290,11 @@
            resize_convolutional_layer(&l, w, h);
        }else if(l.type == MAXPOOL){
            resize_maxpool_layer(&l, w, h);
        }else if(l.type == AVGPOOL){
            resize_avgpool_layer(&l, w, h);
            break;
        }else if(l.type == NORMALIZATION){
            resize_normalization_layer(&l, w, h);
        }else{
            error("Cannot resize this type of layer");
        }