search

Han-Jia / AVIATOR

A practical few-shot learning approach based on MAML.
4 stars 2 forks source link

When I run 'Proto'_ MAML ', an error occurred in the code, #2

Open aosodofogo opened 1 year ago

aosodofogo commented 1 year ago

RuntimeError: One of the differentiated Tensors appears to not have been used in the graph. Set allow_unused=True if this is the desired behavior. The error appears to be due to the calculation "grades=torch. autograd. grad (loss, Tensor_list, create_graph=not first_order)" This problem did not occur when I ran other experiments. I would appreciate it if you could provide a solution

Han-Jia commented 1 year ago

Hello, I updated the file in another project. Could you please send a request to my email yehj@lamda.nju.edu.cn and I'll reply you the new file.

aosodofogo commented 1 year ago

Thank you for your quick response, I have already used my email to Send you a message and look forward to your new file