Bump into some accuracy problem about supervised-reptile HOT 4 CLOSED

For 1: for mini-imagenet, HPs were tuned to maximize validation performance. For omniglot, they we're tuned to maximize training performance, which tended to correlate well with test performance. The CMA hyperparameter optimization code is not included in this repo since it is rather specific to our infrastructure.

For 2: the run scripts in this repo are mainly intended to be used to reproduce our results. If you want to further optimize hyper-parameters, you will want to make sure to only look at the outputs for the validation/training set.

from supervised-reptile.

Did you run the transductive or non-transductive version? The accuracy for transduction is slightly better. Also, note the error margins on these experiments.

from supervised-reptile.

@unixpickle I recap the paper. My mistanke. I thought the accuarcy should be above 90% at least. I am a newbie to meta learning area. I ran the non-transductive version, I will switch to omniglot to try again.Thanks for your quick response!

from supervised-reptile.

@unixpickle Two more question.

1. How did you determine the optimal value of meta_batch_size without any validation process within the outer loop?
2. Why did you include the test set in your training process rather than validation set? It seems a bit odd.

from supervised-reptile.