how to use demo.py to detect new image?

[rw1995]

python demo.py --cfg=./experiments/cfgs/fssd_vgg16_train_voc.yml --demo=./experiments/1.mp4 --type=video

这么打，顺便问一句能加好友吗？一起学习

[wilxy]

python demo.py --cfg=./experiments/cfgs/fssd_vgg16_train_voc.yml --demo=./experiments/1.mp4 --type=video

这么打，顺便问一句能加好友吗？一起学习

你好，我在demo图片的时候遇到了 RuntimeErroe: The size if tensor a(4) must match the size of tensor b(2) at non-singleton dimension 但是我没有改过原始的网络结构啊，请问该怎么解决呢？感谢！

[rw1995]

python demo.py --cfg=./experiments/cfgs/fssd_vgg16_train_voc.yml --demo=./experiments/1.mp4 --type=video 这么打，顺便问一句能加好友吗？一起学习

你好，我在demo图片的时候遇到了 RuntimeErroe: The size if tensor a(4) must match the size of tensor b(2) at non-singleton dimension 但是我没有改过原始的网络结构啊，请问该怎么解决呢？感谢！

抱歉这么晚才看到，请问你解决了吗？

[woodenchild]

@rw1995 hi, 我也遇到这个bug了 Traceback (most recent call last): File ".\demo.py", line 155, in demo_live(args, args.demo_file) File ".\demo.py", line 96, in demo_live _labels, _scores, _coords = object_detector.predict(image) File "D:\Visual_Detect\my_net_detec\ssds.pytorch\lib\ssds.py", line 82, in predict detections = self.detector.forward(out) File "D:\Visual_Detect\my_net_detec\ssds.pytorch\lib\layers\functions\detection.py", line 151, in forward ids, count = nms(boxes, scores, self.nms_thresh, self.top_k) ValueError: not enough values to unpack (expected 2, got 0)

看起来像是没输出的意思，不知道你遇到过没

[robinenrico]

python demo.py --cfg=./experiments/cfgs/fssd_vgg16_train_voc.yml --demo=./experiments/1.mp4 --type=video 这么打，顺便问一句能加好友吗？一起学习

你好，我在demo图片的时候遇到了 RuntimeErroe: The size if tensor a(4) must match the size of tensor b(2) at non-singleton dimension 但是我没有改过原始的网络结构啊，请问该怎么解决呢？感谢！

I encountered the same issue and resolved it by changing the following in the predict function of ssds.py:

Change: scale = torch.Tensor([img.shape[1::-1], img.shape[1::-1]]) To: scale = torch.Tensor([img.shape[1], img.shape[0], img.shape[1], img.shape[0]])

[baoyinhe]

@rw1995 hi, 我也遇到这个bug了 Traceback (most recent call last): File ".\demo.py", line 155, in demo_live(args, args.demo_file) File ".\demo.py", line 96, in demo_live _labels, _scores, _coords = object_detector.predict(image) File "D:\Visual_Detect\my_net_detec\ssds.pytorch\lib\ssds.py", line 82, in predict detections = self.detector.forward(out) File "D:\Visual_Detect\my_net_detec\ssds.pytorch\lib\layers\functions\detection.py", line 151, in forward ids, count = nms(boxes, scores, self.nms_thresh, self.top_k) ValueError: not enough values to unpack (expected 2, got 0)

看起来像是没输出的意思，不知道你遇到过没

+1!请问解决了吗？

[foreverYoungGitHub]

The problem seems be fixed by @robinenrico, will close the issue for now.