edwardzhou130 / PolarSeg

Implementation for PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020)
BSD 3-Clause "New" or "Revised" License
373 stars 80 forks source link

what the set is the pretrained model trained on? #34

Closed LeopoldACC closed 3 years ago

LeopoldACC commented 3 years ago

As the most paper's experiment setting about kitti is that use sequence 00-10 as train set except 08 as validation set?does the pretrained model trained on the whole 00-10 sequence or sequence 00-10 as train set except 08 as validation set? Thks!

edwardzhou130 commented 3 years ago

This is the split of the sequences: https://github.com/edwardzhou130/PolarSeg/blob/d7088e3965e845fb8a563ea759b6c447b65d2506/semantic-kitti.yaml#L186-L211

LeopoldACC commented 3 years ago

that means sequence 08 not used to update the weight of the network,Yeah? Thks!

edwardzhou130 commented 3 years ago

08 will not be used to update the model. It is used to evaluate the performance of the model and decide whether the model should be saved.

LeopoldACC commented 3 years ago

Thks!