About evaluating waymo datasets.

[kx-Z]

I0120 23:13:53.815790 30010 detection_metrics_ops.cc:157] Computing detection metrics for 1026059 predicted boxes. I0120 23:13:53.815812 30010 detection_metrics_ops.cc:159] Parsing prediction [1026059,7][1026059] I0120 23:13:54.045742 30010 detection_metrics_ops.cc:168] Parsing ground truth [1123373,7][1123373] I0120 23:14:19.839358 30010 detection_metrics_ops.cc:221] Done with computing detection metrics.

OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1/AP: 0.7637 OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1/APH: 0.7584 OBJECT_TYPE_TYPE_VEHICLE_LEVEL_2/AP: 0.6738 OBJECT_TYPE_TYPE_VEHICLE_LEVEL_2/APH: 0.6690 OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_1/AP: 0.0000 OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_1/APH: 0.0000 OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_2/AP: 0.0000 OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_2/APH: 0.0000 OBJECT_TYPE_TYPE_SIGN_LEVEL_1/AP: 0.0000 OBJECT_TYPE_TYPE_SIGN_LEVEL_1/APH: 0.0000 OBJECT_TYPE_TYPE_SIGN_LEVEL_2/AP: 0.0000 OBJECT_TYPE_TYPE_SIGN_LEVEL_2/APH: 0.0000 OBJECT_TYPE_TYPE_CYCLIST_LEVEL_1/AP: 0.0000 OBJECT_TYPE_TYPE_CYCLIST_LEVEL_1/APH: 0.0000 OBJECT_TYPE_TYPE_CYCLIST_LEVEL_2/AP: 0.0000 OBJECT_TYPE_TYPE_CYCLIST_LEVEL_2/APH: 0.0000

I just want to experiment with the vehicle class, but it seems to take a lot of time to calculate other classes. How to get rid of them?

[sshaoshuai]

If the prediction dict does not have any predicted pedestrian/cyclist, I guess it will not spend lots of time on calculating the metric of pedestrian./cyclist.

[kx-Z]

Thanks，where to set prediction dict，class names in *.yaml ？

[github-actions[bot]]

This issue is stale because it has been open for 30 days with no activity.

[github-actions[bot]]

This issue was closed because it has been inactive for 14 days since being marked as stale.