分类可以设置为空吗?

[ma062304]

我的实际运用不需要输出种类，可以把model_data中的coco classes和voc classes里的分类清空吗？

[ma062304]

分类文件夹清空后预测显示这个： D:\anaconda\python.exe D:/abc/yolo1/predict.py logs/ep100-loss0.003-val_loss0.003.pth model, anchors, and classes loaded. Input image filename:D:\abc\yolo1\VOCdevkit\VOC2007\JPEGImages\000039.JPG Traceback (most recent call last): File "D:/abc/yolo1/predict.py", line 65, in r_image = yolo.detect_image(image) File "D:\abc\yolo1\yolo.py", line 136, in detect_image outputs = self.bbox_util.decode_box(outputs) File "D:\abc\yolo1\utils\utils_bbox.py", line 109, in decode_box conf.view(batch_size, -1, 1), pred_cls.view(batch_size, -1, self.num_classes)), -1) RuntimeError: cannot reshape tensor of 0 elements into shape [1, -1, 0] because the unspecified dimension size -1 can be any value and is ambiguous

[bubbliiiing]

不可以清空……我建议你看看代码再去掉种类吧

[bubbliiiing]

看看预测过程的代码