RuntimeError：input.size（-1）必须等于input_size

Question

from torch.utils.data import Dataset, DataLoader

if __name__ == '__main__':
x_y_data_set = LyricsDataSet(lyrics)
data_loader = DataLoader(x_y_data_set, batch_size=1,
                         shuffle=True)
trainer.run(data_loader, 10)

RuntimeError                              Traceback (most recent call 
last)
<ipython-input-59-38376625da41> in <module>()
    296     data_loader = DataLoader(x_y_data_set, batch_size=1,
    297                              shuffle=True)
--> 298     trainer.run(data_loader, 10)

14 frames
/usr/local/lib/python3.6/dist-packages/torch/nn/modules/rnn.py in 
check_input(self, input, batch_sizes)
    151             raise RuntimeError(
    152                 'input.size(-1) must be equal to input_size. 
Expected {}, got {}'.format(
--> 153                     self.input_size, input.size(-1)))
    154 
    155     @weak_script_method

RuntimeError: input.size(-1) must be equal to input_size. Expected 300, 
got 96

该代码可以在Anaconda上完美运行，但是当我在Google Colab上运行时，它会给出RuntimeError。我似乎找不到问题。任何帮助将不胜感激。