Lonnnng story short:
When I have my dataloader called dls_test, I can simply run dls_test.decode() and get all the data decoded (de-normalized is mainly what I’m looking for)
When I try and do the same for my same test set inside of my full dataloader:
dls.test_dl.decode() errors out with
*** TypeError: decode() missing 1 required positional argument: ‘b’
I really really really need to decode my test set inside of dls. Any wisdom on what I’m doing wrong?