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
Hi,
I noticed that the label/target is included as a class in the image canvas in your DWTM method. Could this be why your CNN achieves such good results, as it simplifies classification? I've replicated similar results with a basic CNN, possibly due to this inclusion. Could you please clarify if my understanding is correct?
The text was updated successfully, but these errors were encountered:
I didn't notice it during my experiments. If that's the case then yes it's
an issue. Maybe you can try to remove the class and see if you get similar
results. The 3 primary datasets I used did achieve good even with SuperTML
so this would be a good find in a new dataset.
On Wed, Jun 5, 2024, 6:20 PM LAKSHAYA MALETI ***@***.***> wrote:
Hi,
I noticed that the label/target is included as a class in the image canvas
in your DWTM method. Could this be why your CNN achieves such good results,
as it simplifies classification? I've replicated similar results with a
basic CNN, possibly due to this inclusion. Could you please clarify if my
understanding is correct?
—
Reply to this email directly, view it on GitHub
<#2>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AKVABX5IMTVUBXAJQSW76XDZF43GRAVCNFSM6AAAAABI3BYEZOVHI2DSMVQWIX3LMV43ASLTON2WKOZSGMZTMMZTG4ZDIOA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
Hi,
I noticed that the label/target is included as a class in the image canvas in your DWTM method. Could this be why your CNN achieves such good results, as it simplifies classification? I've replicated similar results with a basic CNN, possibly due to this inclusion. Could you please clarify if my understanding is correct?
The text was updated successfully, but these errors were encountered: