maml omniglot - finetuning / test

[shayanfazeli]

Hi,

It appears based on this line that an additional fine-tuning step is needed for a complete MAML example. Is there an example code using higher that has this step?

[brando90]

    # Crucially in our testing procedure here, we do *not* fine-tune
    # the model during testing for simplicity.
    # Most research papers using MAML for this task do an extra
    # stage of fine-tuning here that should be added if you are
    # adapting this code for research.

I am also interested in clarification for this.

[brando90]

@samvelyan does this help? https://github.com/tristandeleu/pytorch-maml

[ligeng0197]

Dont understand what the "fine-tune" mentioned in this annotation represents. In my view, the test procedure is already complete with adapting model to support set of each new task.

[brando90]

related: https://stats.stackexchange.com/questions/550990/what-does-it-mean-to-fine-tune-a-maml-model-for-testing

[brando90]

duplicated: #120