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
Hello everyone, Sorry to bother you.
I am experimenting with the dataset of AVA type. I find that in the file named mmaction\evaluation\metrics\ava_metric.py, existing the code action_thr: float = 0.002, which seems to affect the final AP result. So why should the parameter action_thr be set to 0.002, not 0.5 or 0 or whatever?
My personal thinking is that in multi-label behavior recognition, 0.002 might allow for the capture of as many behavior labels as possible, but why not 0.5? After all, 0.002 is too small. I am not sure whether this idea is correct. Please help me.
Thank you!!
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello everyone, Sorry to bother you.
I am experimenting with the dataset of AVA type. I find that in the file named
mmaction\evaluation\metrics\ava_metric.py
, existing the codeaction_thr: float = 0.002
, which seems to affect the final AP result. So why should the parameter action_thr be set to 0.002, not 0.5 or 0 or whatever?My personal thinking is that in multi-label behavior recognition, 0.002 might allow for the capture of as many behavior labels as possible, but why not 0.5? After all, 0.002 is too small. I am not sure whether this idea is correct. Please help me.
Thank you!!
Beta Was this translation helpful? Give feedback.
All reactions