cenh / citeulike-pytorch Goto Github PK

D:\Edge浏览器下载\CiteULike-PyTorch-master>python main.py
Device: cpu
  0%|                                                                                         | 0/1080 [00:00<?, ?it/s]Skipping token b'2519370' with 1-dimensional vector [b'300']; likely a header
100%|████████████████████████████████████████████████████████████████████████████| 1080/1080 [00:00<00:00, 7714.27it/s]
Traceback (most recent call last):
  File "main.py", line 12, in <module>
    citeulike = citeulike(batch_size=200)
  File "D:\Edge浏览器下载\CiteULike-PyTorch-master\data.py", line 65, in __init__
    self.doc_title.build_vocab(self.train_set, max_size=None, vectors=vocab.Vectors('wiki.en.vec', url=url))
  File "D:\learn\Anaconda3\lib\site-packages\torchtext\vocab\vectors.py", line 59, in __init__
    self.cache(name, cache, url=url, max_vectors=max_vectors)
  File "D:\learn\Anaconda3\lib\site-packages\torchtext\vocab\vectors.py", line 142, in cache
    dim))
RuntimeError: Vector for token b'money' has 81 dimensions, but previously read vectors have 300 dimensions. All vectors must have the same number of dimensions.