search

Thinklab-SJTU / Bench2Drive

[NeurIPS 2024 Datasets and Benchmarks Track] Closed-Loop E2E-AD Benchmark Enhanced by World Model RL Expert
Other
1.3k stars 85 forks source link

Failed when run evaluation #33

Closed donkehuang closed 4 months ago

donkehuang commented 4 months ago

Hi, thans for your great work and effort. I have trouble when run the evaluation of VAD and here are the command i used and error info. image `bash leaderboard/scripts/run_evaluation_debug.sh

`

when i run the carla seperately, it works and shows the figure like this image

And my evaluation config is like this, are these config right for VAD evaluation? `#!/bin/bash BASE_PORT=30000 BASE_TM_PORT=50000 IS_BENCH2DRIVE=True BASE_ROUTES=leaderboard/data/bench2drive220 TEAM_AGENT=leaderboard/team_code/vad_b2d_agent.py TEAM_CONFIG=~/Bench2Drive/Bench2DriveZoo/ckpts/vad_b2d_base.pth # for UniAD and VAD BASE_CHECKPOINT_ENDPOINT=eval SAVE_PATH=~/Bench2DriveZoo/test/ PLANNER_TYPE=only_traj

GPU_RANK=3 PORT=$BASE_PORT TM_PORT=$BASE_TM_PORT ROUTES="${BASE_ROUTES}.xml" CHECKPOINT_ENDPOINT="${BASE_CHECKPOINT_ENDPOINT}.json" bash leaderboard/scripts/run_evaluation.sh $PORT $TM_PORT $IS_BENCH2DRIVE $ROUTES $TEAM_AGENT $TEAM_CONFIG $CHECKPOINT_ENDPOINT $SAVE_PATH $PLANNER_TYPE $GPU_RANK `

Thanks a lot!

jayyoung0802 commented 4 months ago

It seems that there is a problem with the carla version. Please check the carla version(0.9.15).