demo.ipynb執行Forward pass 會出現 ValueError: not enough values to unpack (expected 2, got 0) - Cupoy

demo.ipynb 內有一段'''要先用Variable包裝才能送給Pytorch模型'''xx ...

cvdl-1,cvdl-1-d29

2020/01/08 下午 06:24

電腦視覺深度學習討論版

林弘敏

觀看數：10

回答數：1

收藏數：1

cvdl-1

cvdl-1-d29

demo.ipynb 內有一段

'''要先用Variable包裝才能送給Pytorch模型'''

xx = Variable(x.unsqueeze(0)) # wrap tensor in Variable

if torch.cuda.is_available():

xx = xx.cuda()

'''Forward Pass'''

y = net(xx)

執行後會出現錯誤, 內容如下:

ValueError Traceback (most recent call last)

4 xx = xx.cuda()

5 '''Forward Pass'''

----> 6 y = net(xx)

D:\Anaconda3\lib\site-packages\torch\nn\modules\module.py in __call__(self, *input, **kwargs)

475 result = self._slow_forward(*input, **kwargs)

476 else:

--> 477 result = self.forward(*input, **kwargs)

478 for hook in self._forward_hooks.values():

479 hook_result = hook(self, input, result)

~\ssd.py in forward(self, x)

101 self.softmax(conf.view(conf.size(0), -1,

102 self.num_classes)), # conf preds

--> 103 self.priors.type(type(x.data)) # default boxes

104 )

105 else:

~\layers\functions\detection.py in forward(self, loc_data, conf_data, prior_data)

52 boxes = decoded_boxes[l_mask].view(-1, 4)

53 # idx of highest scoring and non-overlapping boxes per class

---> 54 ids, count = nms(boxes, scores, self.nms_thresh, self.top_k)

55 output[i, cl, :count] = \

56 torch.cat((scores[ids[:count]].unsqueeze(1),

ValueError: not enough values to unpack (expected 2, got 0)