Fixed multi-GPU training for Tensor Cores

AlexeyAB

2018-03-09 880cf187d87c904f5fe574802ecff99118643f2d

 src/detection_layer.c

@@ -58,8 +58,8 @@
            int index = b*l.inputs;
            for (i = 0; i < locations; ++i) {
                int offset = i*l.classes;
                softmax_array(l.output + index + offset, l.classes, 1,
                        l.output + index + offset);
                softmax(l.output + index + offset, l.classes, 1,
                        l.output + index + offset, 1);
            }
        }
    }