search

xinge008 / Cylinder3D

Rank 1st in the leaderboard of SemanticKITTI semantic segmentation (both single-scan and multi-scan) (Nov. 2020) (CVPR2021 Oral)
Apache License 2.0
859 stars 180 forks source link

expected input batch_size (6) to match target batch_size (2) #58

Closed caoyifeng001 closed 3 years ago

caoyifeng001 commented 3 years ago

when i use batch size = 6 for train get error image

Colin97 commented 3 years ago

Hi @caoyifeng001 , How do you solve this?

caoyifeng001 commented 3 years ago

训练集最后一组数据不到batch size的大小，只需要predict_labels = my_model(val_pt_fea_ten, val_grid_ten, val_batch_size) 改成 predict_labels = my_model(val_pt_fea_ten, val_grid_ten, val_label_tensor.shape[0]) 就可以了。