Test Time Augmentation throwing a value error

Question

I am trying to use the Test Time Augmentation on my Classifier:

log_preds,y = learn.TTA(scale=1.1, ds_type=DatasetType.Valid, with_loss=True)

And this is the error that it threw:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-14-f33d9737819a> in <module>()
----> 1 log_preds,y = learn.TTA(scale=1.1, ds_type=DatasetType.Valid, with_loss=True)
      2 probs = np.mean(np.exp(log_preds),0)
      3 
      4 accuracy(probs, y)

ValueError: too many values to unpack (expected 2)

Initially, I tried to find a way to use my custom test set in TTA but couldn’t figure out how to do it and DatasetType.Test was throwing an error so I decided to go with DatasetType.Valid and after running 8 epochs, I got the above error.

Answer 1

The error message indicate that learn.TTA returns more than 2 values,but you only got log_predsand y for them. You want to find out what exactly does learn.TTA return