nlp-beginner_solution
nlp-beginner_solution copied to clipboard
task2的dataset失效了
不知道是不是因为时效的原因, task2
的 dataset
和文件名不完全匹配, 可以修改一下
在我把文件名改成 train.tsv
和 test.tsv
之后
运行这段代码的时候
#构建词典,字符映射到embedding
#TEXT.vocab.vectors 就是词向量
TEXT.build_vocab(train_data, vectors= 'glove.6B.50d', #可以提前下载好
unk_init= lambda x:torch.nn.init.uniform_(x, a=-0.25, b=0.25))
LABEL.build_vocab(train_data)
#得到索引,PAD_TOKEN='<pad>'
PAD_INDEX = TEXT.vocab.stoi[PAD_TOKEN]
TEXT.vocab.vectors[PAD_INDEX] = 0.0
colab上提示这个
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-58-88fe1b0b4be6> in <module>()
3
4 TEXT.build_vocab(train_data, vectors= 'glove.6B.50d', #可以提前下载好
----> 5 unk_init= lambda x:torch.nn.init.uniform_(x, a=-0.25, b=0.25))
6 LABEL.build_vocab(train_data)
7
1 frames
/usr/local/lib/python3.7/dist-packages/torchtext/legacy/data/field.py in build_vocab(self, *args, **kwargs)
290 sources.append(arg)
291 for data in sources:
--> 292 for x in data:
293 if not self.sequential:
294 x = [x]
/usr/local/lib/python3.7/dist-packages/torchtext/legacy/data/dataset.py in __getattr__(self, attr)
152 if attr in self.fields:
153 for x in self.examples:
--> 154 yield getattr(x, attr)
155
156 @classmethod
AttributeError: 'Example' object has no attribute 'Phrase'
不知道是什么原因