Jiazhi Yang

@181860153 It will take a lot of engineering effort to adapt the codebase to other datasets, as it's heavily coupled with nuscenes annotations for multiple tasks throughout the data process,...

Hi @wljungbergh, thank you so much for sharing this valuable information. Could you please make a pull request for the code change? I'd like to present your verification in the...

@yudefu Hi, both of your installed mmcv and mmdet3d have higher version than what's required in this project. You may need to create a new environment and install the requisite...

Hi @wljungbergh ! I prefer it to be configurable. How about distinguishing the two evaluation protocols by the paper name? One can specify the evaluation strategy using `evaluation_strategy='stp3' or 'uniad'`...

I will update the README and other instruction files accordingly. No need to worry about it.

Also, I want to know whether the misalignment issue also happens on the `collision rate` metric. Is the stp3 metric cumulative while uniad is non-cumulative? Have you looked into that...

Understood. After modifying the code, could you please make a complete table for both L2 and CR metrics under the `stp3` and `uniad` strategies like the one you posted? It...

Great! Thanks for your help.

Hi @thunguyenth. To terminate the process in Linux, you can use the command `pkill -9 python`.

@haichen-ber Hi, does these values come from evaluation? Could you please specify this problem a little bit? Any log information or screenshot would be helpful.