search

autonomousvision / transfuser

[PAMI'23] TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving; [CVPR'21] Multi-Modal Fusion Transformer for End-to-End Autonomous Driving
MIT License
1.15k stars 189 forks source link

How long did the training take? #49

Closed Watson52 closed 2 years ago

Watson52 commented 2 years ago

Thank you for the open source code and dataset generation methods. I would like to ask how long did the training take？

ap229997 commented 2 years ago

Geometric Fusion and TransFuser take ~2 days, Late Fusion takes 1 day, CILRS and AIM complete easily within a day and LBC takes 2-3 days since it involves training both the teacher and student models. All these estimates are using a single 1080Ti GPU. The training time can also depend on where you are loading the data from. In our case, all the data was stored on the SSD of the GPU node so dataloading was pretty fast. If your data storage has heavy I/O then that can also affect the training time.

Watson52 commented 2 years ago

I got it. Thank you for your quick reply!