search

bradyz / cross_view_transformers

Cross-view Transformers for real-time Map-view Semantic Segmentation (CVPR 2022 Oral)
MIT License
525 stars 80 forks source link

Evaluation code #5

Closed xiaoyucheri closed 2 years ago

xiaoyucheri commented 2 years ago

Hi, Thanks for your great work and sharing the code! I didn't find the codes for evaluation, will you release it sooner? Thanks!

babyporgrammer commented 2 years ago

thanks for the code, when will we have the eval code?

sparkingcoding commented 2 years ago

Since the nuScenes segmentation is performed on the validation set, I saw a lot of repos that don't have evaluation codes and just use the validation evaluation curve result during training. I tried to train the CVT model, and after it was finished, I checked the wanddb board and found the best result on validation@iou0.5 curve is around 36.0. I guess this is how it estimates the result,

xiaoyucheri commented 2 years ago

@sparkingcoding makes sense, thanks