validation on custom lidar only dataset (indoor_eval)

[gynaa]

Hello,

I have a question regarding the validation step. I am trying to train PartA2 on a custom lidar only dataset. The training works fine. However, during the validation my AP is always 0.0000. I was wondering does the evaluation in indoor_eval require images to work properly or should it also work with a lidar only dataset?

[sunjiahao1999]

Indoor_eval can work with lidar-only dataset. Perhaps it may be because your pred_label id mismatch your gt_label id?

[gynaa]

thank you very much for your quick response. so i just realized that dt_annos is filled with empty entires. meaning that pred is completely empty. I checked the forward training functions and in there bounding box predicitions are being made. Any idea to why no predictions are being made in the validation?

Here is my validation pipeline, if that helps:
eval_pipeline = [ dict(type='LoadPointsFromFile', coord_type='LIDAR', load_dim=4, use_dim=3), dict(type='MultiScaleFlipAug3D', img_scale=(1, 1), pts_scale_ratio=1, flip=False, transforms=[ dict(type='GlobalRotScaleTrans', rot_range=[0, 0], scale_ratio_range=[1., 1.], translation_std=[0, 0, 0]), dict(type='PointsRangeFilter', point_cloud_range=point_cloud_range), dict(type='DefaultFormatBundle3D', class_names=classes, with_label=False), dict(type='Collect3D', keys=['points'])])]

[gynaa]

any help would be appreciated :)