You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Can I ask you about the torch implementation version, in your pytorch implementation,the penal was directly added to entropyloss and loss.backward(),in torch implementation,what should i do to implement
the similar function,should icreate a new class inherited from nn.Criterion to implement it?
thank you for Implementation of A Structured Self-attentive Sentence Embedding,Do you have the torch
implementation version
The text was updated successfully, but these errors were encountered: