search

usail-hkust / LLMTSCS

Official code for article "LLMLight: Large Language Models as Traffic Signal Control Agents".
139 stars 12 forks source link

Error when run Policy Refinement Data Collection #16

Open DA21S321D opened 2 months ago

DA21S321D commented 2 months ago

截图 2024-04-20 12-50-26 When i run python run_policy_refinement_data_collection.py --llm_model dataCollect7BFT --llm_path /root/autodl-tmp/HuggingFace-Download-Accelerator/modelDownload/merged --dataset jinan --traffic_file anon_3_4_jinan_real.json

Gungnir2099 commented 2 months ago

You are using the action-value network of Advanced-Colight trained in Hangzhou to collect the policy refinement data for fine-tuning LightGPT for Jinan. Please either train a new action-value network in Jinan or collect the refinement data for Hangzhou.