You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In your paper, your specified that "Target only" refer on training a model on the target domain from scratch on labeled images.
Specifically on domain C, the training samples are not annotated, that out of 150 samples in domain C, 50 is not labeled.
So the question is:
For the target only on domain C, did you split only the labeled samples (50) to train valid and test for training then model?
And
For the IPLC method on domain C, did you split all the samples to train valid and test but in a way that the 50 unlabled samples will be on the train set?
Thanks
Ido.
The text was updated successfully, but these errors were encountered:
"Hello, I’m also researching this paper recently. I’d like to ask if you have reproduced its code. I feel like there are quite a few issues in the code."
Hi,
Thanks for your great work.
In your paper, your specified that "Target only" refer on training a model on the target domain from scratch on labeled images.
Specifically on domain C, the training samples are not annotated, that out of 150 samples in domain C, 50 is not labeled.
So the question is:
For the target only on domain C, did you split only the labeled samples (50) to train valid and test for training then model?
And
For the IPLC method on domain C, did you split all the samples to train valid and test but in a way that the 50 unlabled samples will be on the train set?
Thanks
Ido.
The text was updated successfully, but these errors were encountered: