yanwii / seq2seq

基于Pytorch的中文聊天机器人 集成BeamSearch算法
Apache License 2.0
241 stars 87 forks source link

expected tensor [5798 x 100] and src [600 x 100] #5

Closed skarl-api closed 7 years ago

skarl-api commented 7 years ago

这个是什么错误 好奇怪的错误 While copying the parameter named encoder.embedding.weight, whose dimensions in the model are torch.Size([5798, 100]) and whose dimensions in the checkpoint are torch.Size([600, 100]), ... inconsistent tensor size, expected tensor [5798 x 100] and src [600 x 100] to have the same number of elements, but got 579800 and 60000 elements respectively

skarl-api commented 7 years ago

@yanwii 我猜测 是 params.pkl原有的训练是600 但是这次想训练 5798 导致错误 但是 如何初始化一个空白的 params.pkl呢?