Any suggestions for improving the effectiveness of the model, such as increasing the number of model parameters?

tusen-ai / SST

Code for a series of work in LiDAR perception, including SST (CVPR 22), FSD (NeurIPS 22), FSD++ (TPAMI 23), FSDv2, and CTRL (ICCV 23, oral).

Apache License 2.0

801 stars 102 forks source link

Any suggestions for improving the effectiveness of the model, such as increasing the number of model parameters? #183

Open 2anchao opened 9 months ago

2anchao commented 9 months ago

hello author: I find that fsdv2 model training occupy less memory, config is configs/fsdv2/fsdv2_nusc_2x.py, and I want use fsdv2 for autolabel, so do you have any suggestions on the model structure to improve the model, such as backbone choice or other module?