I met an Error during evaluation. How can I solve it?

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

1.12k stars 186 forks source link

I met an Error during evaluation. How can I solve it? #215

Closed jingzhengli closed 4 months ago

jingzhengli commented 4 months ago

========= Preparing RouteScenario33 (repetition 0) ========= >_ Setting up the agent

Could not set up the required agent: > module 'transfuser' has no attribute 'get_entry_point'

Traceback (most recent call last): File "/home/ljzdisk/ad/transfuser/leaderboard/leaderboard/leaderboard_evaluator_local.py", line 272, in _load_and_run_scenario agent_class_name = getattr(self.module_agent, 'get_entry_point')() AttributeError: module 'transfuser' has no attribute 'get_entrypoint' >_ Registering the route statistics

jingzhengli commented 4 months ago

========= Preparing RouteScenario33 (repetition 0) ========= >_ Setting up the agent

Could not set up the required agent: > module 'transfuser' has no attribute 'get_entry_point'

Traceback (most recent call last): File "/home/ljzdisk/ad/transfuser/leaderboard/leaderboard/leaderboard_evaluator_local.py", line 272, in _load_and_run_scenario agent_class_name = getattr(self.module_agent, 'get_entry_point')() AttributeError: module 'transfuser' has no attribute 'get_entrypoint' >_ Registering the route statistics

It is not clearly how to set "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py" in local_evaluation.sh I set to "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/transfuser.py"

jingzhengli commented 4 months ago

========= Preparing RouteScenario33 (repetition 0) ========= > Setting up the agent Could not set up the required agent: >_ module 'transfuser' has no attribute 'get_entry_point' Traceback (most recent call last): File "/home/ljzdisk/ad/transfuser/leaderboard/leaderboard/leaderboard_evaluator_local.py", line 272, in _load_and_run_scenario agent_class_name = getattr(self.module_agent, 'get_entry_point')() AttributeError: module 'transfuser' has no attribute 'get_entrypoint' >_ Registering the route statistics

It is not clearly how to set "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py" in local_evaluation.sh I set to "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/transfuser.py"

Then, It doesn't have any error, but report this imformation.

> Registering the global statistics

Kait0 commented 4 months ago

what is not clear? You don't need to change this. submission_agent.py is the correct file. The transfuser.py file only contains the transfuser backbone, there are other files involved.

Registering the global statistics

This can happen if your code crashed before or so. Try deleting the ${WORK_DIR}/results/transfuser_longest6.json file and try again (might reveal the real error).

jingzhengli commented 4 months ago

what is not clear? You don't need to change this. submission_agent.py is the correct file. The transfuser.py file only contains the transfuser backbone, there are other files involved.

Registering the global statistics

This can happen if your code crashed before or so. Try deleting the ${WORK_DIR}/results/transfuser_longest6.json file and try again (might reveal the real error).

Thanks for your reply. If I use this comment: "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py". I will met an error like this： ./leaderboard/scripts/local_evaluation.sh /home/ljzdisk/carla /home/ljzdisk/ad/transfuser ./leaderboard/scripts/local_evaluation.sh: line 38: 69613 Segmentation fault (core dumped) python3 ${LEADERBOARD_ROOT}/leaderboard/leaderboard_evaluator_local.py --scenarios=${SCENARIOS} --routes=${ROUTES} --repetitions=${REPETITIONS} --track=${CHALLENGE_TRACK_CODENAME} --checkpoint=${CHECKPOINT_ENDPOINT} --agent=${TEAM_AGENT} --agent-config=${TEAM_CONFIG} --debug=${DEBUG_CHALLENGE} --resume=${RESUME} I cannot know where is the error. Thanks again. I want to know if the error come from the version of carla==0.9.15 that I installed early.

jingzhengli commented 4 months ago

what is not clear? You don't need to change this. submission_agent.py is the correct file. The transfuser.py file only contains the transfuser backbone, there are other files involved.

Registering the global statistics

This can happen if your code crashed before or so. Try deleting the ${WORK_DIR}/results/transfuser_longest6.json file and try again (might reveal the real error).

Thanks for your reply. If I use this comment: "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py". I will met an error like this： ./leaderboard/scripts/local_evaluation.sh /home/ljzdisk/carla /home/ljzdisk/ad/transfuser ./leaderboard/scripts/local_evaluation.sh: line 38: 69613 Segmentation fault (core dumped) python3 ${LEADERBOARD_ROOT}/leaderboard/leaderboard_evaluator_local.py --scenarios=${SCENARIOS} --routes=${ROUTES} --repetitions=${REPETITIONS} --track=${CHALLENGE_TRACK_CODENAME} --checkpoint=${CHECKPOINT_ENDPOINT} --agent=${TEAM_AGENT} --agent-config=${TEAM_CONFIG} --debug=${DEBUG_CHALLENGE} --resume=${RESUME} I cannot know where is the error. Thanks again. I want to know if the error come from the version of carla==0.9.15 that I installed early.

Now, I know the error comes form "self.module_agent = importlib.import_module(module_name)" in leaderboard_evaluator_local.py. But I don't know why it happens and how to solve it. Thanks again,

jingzhengli commented 4 months ago

what is not clear? You don't need to change this. submission_agent.py is the correct file. The transfuser.py file only contains the transfuser backbone, there are other files involved.

Registering the global statistics

This can happen if your code crashed before or so. Try deleting the ${WORK_DIR}/results/transfuser_longest6.json file and try again (might reveal the real error).

Thanks for your reply. If I use this comment: "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py". I will met an error like this： ./leaderboard/scripts/local_evaluation.sh /home/ljzdisk/carla /home/ljzdisk/ad/transfuser ./leaderboard/scripts/local_evaluation.sh: line 38: 69613 Segmentation fault (core dumped) python3 ${LEADERBOARD_ROOT}/leaderboard/leaderboard_evaluator_local.py --scenarios=${SCENARIOS} --routes=${ROUTES} --repetitions=${REPETITIONS} --track=${CHALLENGE_TRACK_CODENAME} --checkpoint=${CHECKPOINT_ENDPOINT} --agent=${TEAM_AGENT} --agent-config=${TEAM_CONFIG} --debug=${DEBUG_CHALLENGE} --resume=${RESUME} I cannot know where is the error. Thanks again. I want to know if the error come from the version of carla==0.9.15 that I installed early.

Now, I know the error comes form "self.module_agent = importlib.import_module(module_name)" in leaderboard_evaluator_local.py. But I don't know why it happens and how to solve it. Thanks again,

The same error as previous issue "https://github.com/autonomousvision/transfuser/issues/200".

Kait0 commented 4 months ago

This codebase is not compatible with carla 0.9.15 (You need leaderboard 2.0 for that and as far as I know there are no public codebases available for that yet). You need carla version 0.9.10.1 to run the code. The setup script will download it for you.

jingzhengli commented 4 months ago

what is not clear? You don't need to change this. submission_agent.py is the correct file. The transfuser.py file only contains the transfuser backbone, there are other files involved.

Registering the global statistics

This can happen if your code crashed before or so. Try deleting the ${WORK_DIR}/results/transfuser_longest6.json file and try again (might reveal the real error).

Thanks for your reply. If I use this comment: "export TEAM_AGENT=${WORK_DIR}/team_code_transfuser/submission_agent.py". I will met an error like this： ./leaderboard/scripts/local_evaluation.sh /home/ljzdisk/carla /home/ljzdisk/ad/transfuser ./leaderboard/scripts/local_evaluation.sh: line 38: 69613 Segmentation fault (core dumped) python3 ${LEADERBOARD_ROOT}/leaderboard/leaderboard_evaluator_local.py --scenarios=${SCENARIOS} --routes=${ROUTES} --repetitions=${REPETITIONS} --track=${CHALLENGE_TRACK_CODENAME} --checkpoint=${CHECKPOINT_ENDPOINT} --agent=${TEAM_AGENT} --agent-config=${TEAM_CONFIG} --debug=${DEBUG_CHALLENGE} --resume=${RESUME} I cannot know where is the error. Thanks again. I want to know if the error come from the version of carla==0.9.15 that I installed early.

Now, I know the error comes form "self.module_agent = importlib.import_module(module_name)" in leaderboard_evaluator_local.py. But I don't know why it happens and how to solve it. Thanks again,

The same error as previous issue "#200".

This codebase is not compatible with carla 0.9.15 (You need leaderboard 2.0 for that and as far as I know there are no public codebases available for that yet). You need carla version 0.9.10.1 to run the code. The setup script will download it for you.

Many thanks.