6.5.2 RNN简洁实现,使用相邻采样时,state在新的epoch中未被初始化 #161
Comments
jianli-Alex
changed the title
jianli-Alex
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
bug描述
state = None放在了for epoch in range(num_epochs)上. 因此,在新的epoch中,state实际上没有被初始化,而是使用上一epoch最后一次迭代最后一个时间步的state, 这样并不合理.要把
state = None放在了for epoch in range(num_epochs)下面,for X, Y in data_iter上面版本信息
pytorch: 1.6.0
torchvision: 0.7.0
torchtext:
...
The text was updated successfully, but these errors were encountered: