Runtime Error in loss.backward()

[AayushGrover]

Hello,

I was trying to run the imaml codebase. However, I get an error while training the learner class object. The error message is as follows - "RuntimeError: Trying to backward through the graph a second time, but the buffers have already been freed. Specify retain_graph=True when calling backward the first time." This is happening in the loss.backward() mentioned in the learn_on_data function of the Learner class. Kindly let me know how this error can be fixed.