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
When exploring the dataset, it's common to adjust the range of the pose results returned by filter_poses. Currently if you want to modify the default pose query range, you have to do this:
To support the common case, it would be very convenient to support a named array argument to filter_poses(range=[min, max, stride]);
If you are worried about making it too easy to hang the server, add some error checking into filter_poses that errors out if the query requests a set of poses that is pretty large. (e.g., more than 2M).
The text was updated successfully, but these errors were encountered:
When exploring the dataset, it's common to adjust the range of the pose results returned by filter_poses. Currently if you want to modify the default pose query range, you have to do this:
filtered = filter_poses('pose', hands_above_head, [Pose.LWrist, Pose.Nose, Pose.RWrist], Pose.objecfts.all().order_by('id').select_related('frame', 'frame__video')[0:250000:10])
To support the common case, it would be very convenient to support a named array argument to filter_poses(range=[min, max, stride]);
If you are worried about making it too easy to hang the server, add some error checking into filter_poses that errors out if the query requests a set of poses that is pretty large. (e.g., more than 2M).
The text was updated successfully, but these errors were encountered: