怎么解决这个问题，多谢！先感谢分享

[joyzyyz]

(base) D:\F\VideoCaption19>python videoCaption.py video10034.mp4
in_file: video10034.mp4
IMG_SIZE: (256, 192)
--processing video...
18%|█████████████████████████████████▋ |
40%|████████████████████████████████████████████████████████████████████████▊
62%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
84%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████████████| 527/527 [00:00<00:00, 1125.99it/s]

--processing key frames...
0%| | 0/15 [00:00<?, ?it/s]D
:\F\VideoCaption19\img2txt.py:99: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
img_feats = feat_extractor(Variable(img, volatile=True))
D:\F\VideoCaption19\model.py:60: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
img = t.autograd.Variable(img.unsqueeze(0), volatile=True)
D:\F\VideoCaption19\utils\beam_search.py:156: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.
logprobs = log_softmax(output)
D:\F\VideoCaption19\utils\beam_search.py:180: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
input_feed = Variable(input_feed, volatile=True)
0%| | 0/15 [00:00<?, ?it/s]
Traceback (most recent call last):
File "videoCaption.py", line 121, in
video2txt(in_file)
File "videoCaption.py", line 73, in video2txt
txts += generate_txt(img, tr4s, resnet50, cap_model, opt.use_gpu)
File "D:\F\VideoCaption19\img2txt.py", line 105, in generate_txt
results = model.generate(img_feats.data[0])
File "D:\F\VideoCaption19\model.py", line 64, in generate
for sent in sentences]
File "D:\F\VideoCaption19\model.py", line 64, in
for sent in sentences]
File "D:\F\VideoCaption19\model.py", line 63, in
sentences = [' '.join([self.ix2word[idx] for idx in sent])
KeyError: tensor(51)

[zhuchiheng]

(base) D:\F\VideoCaption19>python videoCaption.py video10034.mp4
in_file: video10034.mp4
IMG_SIZE: (256, 192)
--processing video...
18%|█████████████████████████████████▋ |
40%|████████████████████████████████████████████████████████████████████████▊
62%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
84%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████████████| 527/527 [00:00<00:00, 1125.99it/s]

--processing key frames...
0%| | 0/15 [00:00<?, ?it/s]D
:\F\VideoCaption19\img2txt.py:99: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
img_feats = feat_extractor(Variable(img, volatile=True))
D:\F\VideoCaption19\model.py:60: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
img = t.autograd.Variable(img.unsqueeze(0), volatile=True)
D:\F\VideoCaption19\utils\beam_search.py:156: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.
logprobs = log_softmax(output)
D:\F\VideoCaption19\utils\beam_search.py:180: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
input_feed = Variable(input_feed, volatile=True)
0%| | 0/15 [00:00<?, ?it/s]
Traceback (most recent call last):
File "videoCaption.py", line 121, in
video2txt(in_file)
File "videoCaption.py", line 73, in video2txt
txts += generate_txt(img, tr4s, resnet50, cap_model, opt.use_gpu)
File "D:\F\VideoCaption19\img2txt.py", line 105, in generate_txt
results = model.generate(img_feats.data[0])
File "D:\F\VideoCaption19\model.py", line 64, in generate
for sent in sentences]
File "D:\F\VideoCaption19\model.py", line 64, in
for sent in sentences]
File "D:\F\VideoCaption19\model.py", line 63, in
sentences = [' '.join([self.ix2word[idx] for idx in sent])
KeyError: tensor(51)

@joyzyyz
改为:
sentences = [' '.join([self.ix2word[idx.item()] for idx in sent])