search

weiyithu / SurroundOcc

[ICCV 2023] SurroundOcc: Multi-camera 3D Occupancy Prediction for Autonomous Driving
Apache License 2.0
826 stars 108 forks source link

Training with my own data. #92

Open mlxgzx opened 11 months ago

mlxgzx commented 11 months ago

To train a new model with my own data, what else do I need other than the gt_occupancy obtain from process_your_own_data.py??

weiyithu commented 10 months ago

Hi, you need to generate new pkl with your own data. For nuScenes, we generate pkl by python tools/create_data.py nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag nuscenes --version v1.0 --canbus ./data, which is same with BEVFormer instructions.

HelloYongge commented 5 months ago

I have generate gt_xxx.npy with process_your_own_data.py, if we generate pkl data with nuscenes create_data.py, how to invovle the gt.npy in pkl files?