Minor fix for multi-GPU training with -map flag

pull/2111/head
AlexeyAB 6 years ago
parent 06232b31f4
commit b87080882a
  1. 4
      src/detector.c

@ -81,9 +81,10 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
list *plist = get_paths(train_images);
train_images_num = plist->size;
free_list(plist);
cuda_set_device(net.gpu_index);
printf(" Prepare additional network for mAP calculation...\n");
net_map = parse_network_cfg_custom(cfgfile, 1);
int k;
for (k = 0; k < net.n; ++k) {
layer l = net.layers[k];
@ -105,7 +106,6 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
}
}
const int actual_batch_size = net.batch * net.subdivisions;
if (actual_batch_size == 1) {
printf("\n Error: You set incorrect value batch=1 for Training! You should set batch=64 subdivision=64 \n");

Loading…
Cancel
Save