Hi,
I'm fairly new to deep learning and I recently downloaded and followed the mmdetection3d documentation instructions on how to prepare the nuscenes v1.0-mini dataset. I was able to train the model using the centrepoint configuration file : centerpoint_pillar02_second_secfpn_8xb4-cyclic-20e_nus-3d and ran it one gpu. It completed but towards the end, when evaluating the model. I got the following error "AssertionError: Samples in split doesn't match samples in predictions.".
The error image
Hi, I'm fairly new to deep learning and I recently downloaded and followed the mmdetection3d documentation instructions on how to prepare the nuscenes v1.0-mini dataset. I was able to train the model using the centrepoint configuration file : centerpoint_pillar02_second_secfpn_8xb4-cyclic-20e_nus-3d and ran it one gpu. It completed but towards the end, when evaluating the model. I got the following error "AssertionError: Samples in split doesn't match samples in predictions.".
The error image
How I structured the dataset
I would appreciate any pointers. Thanks!