Open ZQ2413262560 opened 1 year ago
@ZQ2413262560 could you please provide me the basic info of your team: participants names, affiliation, level of education (BSc, MSc, PhD), team name, contact email
Note, requires
conda install opencv
@ZQ2413262560 could you please provide me the basic info of your team: participants names, affiliation, level of education (BSc, MSc, PhD), team name, contact email participants names: Qin Zhang, Linrui Zhang,Haoyu Wang,Zichen Yan, Shoujie Li,Xueqian Wang,Li Shen. affiliation: Tsinghua University, Beijing, China. JD ExploreAcademy, Beijing, China. level of education: MSc Team Name: AIRobotLab contact email: zhangqin21@mails.tsinghua.edu.cn
Our team is the same as the one registered using zhanglr.auto@gmail.com
before. Please refer to the current information. Thank you!
@ZQ2413262560 I now launched the scripts that will collect the simulation results for all submissions: the 7-8 hours you mentioned are per-level or for all 4 levels (level0.yaml
to level3.yaml
) combined? In case time didn't allow to run and score all submissions by Sunday, do you have code to run single-episodes run from trained models?
@JacopoPan The code I've submitted now is the code for testing.
You can run it directly with the codepython3 getting_start.py --overrides./levelx.yaml
, but you may need to change the level0.yaml
to level3.yaml
configuration file because in our setting we set the done_on_violation
and done_on_collision
to False
@JacopoPan Hi Jacopo! Code runs using commands:
cd competition/
python3 getting_started.py --overrides level0.yaml
It takes about 7-8 hours to train on my machine (1 RTX3090).Time is limited, and some ideas are too late to put into practice. If there is a chance, I will continue to try this task! I would like to thank the organizers for this opportunity, Thank you!