Open plyu3 opened 1 year ago
Hi, Did you freezed/added any layer while applying transfer learning from CMIP base model to SODA data?
Hi @plyu3,
We set the entire duration of each dataset (i.e., CMIP5, SODA, and GODAS) as the reference time and calculated the anomaly.
Hi @sreerajfab,
No, we did not freeze layers or add new layers in the transfer learning process.
Hello,
I was impressed by such a nice paper. I want to know how is your anomaly calculated? What is your reference time for CMIP5, SODA and GODAS?
Thank you!