Validation accuracy stored in learn.interpret?

I’ve been able to get accuracy from

preds, y = learn.get_preds()

But is there a way to extract the validation accuracy from an interp object? Can anyone point me to more info on everything that interp does? The docs feel a little sparse, and I’ve looked at the first few lessons.

1 Like

Getting closer. Shift tab of interp. reveals y_true and pred_class functions. But you can’t plug them into the accuracy() :frowning: