1 line
12 KiB
JSON
1 line
12 KiB
JSON
{"repo": "adamwilliam8/myproject1", "files": {"reproduce1/config.yaml": {"63c3c68d1df09c25a68e0c4fe5b97af3481d7041": {"t": 1785746312.3231523, "n": 3411, "h": "c30456f9f5ee2db0", "c": "### 如环境变动要自己去环境仓库 看需要导入哪些相关文件\n\n# YOUR LLMs API CONFIG\n### DEEPSEEK\nDEEPSEEK_KEY: \"sk-1495882192234e668f470fba3e39f9ce\"\nDEEPSEEK_MODEL: \"deepseek-reasoner\"\nDEEPSEEK_BASE_URL: \"https://api.deepseek.com\"\nDEEPSEEK_MAX_TOKENS: 64000 # 8192 最大输出tokens\nDEEPSEEK_TEMPERATURE: 0.3 # reasoner用不了温度\n### qwen\nQWEN_KEY: \"sk-1c71165453eb4ad192753babda9bffe0\"\nQWEN_MODEL: \"qwen3-coder-plus\"\nQWEN_BASE_URL: \"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nQWEN_MAX_INPUT_TOKENS: 997,952 # 最大输入tokens\nQWEN_MAX_TOKENS: 65,536 # 最大输出tokens\nQWEN_ENABLE_THINKING: True # 是否开启思考模式\n\n### ENV FILE ADDRESS\nENV_FILE_ADDRESS: \"../HighwayEnv-RL_env/highway_env\" # your path to the all env file\nENV_ID: \"highway-v0\"\nMAX_EPISODE_STEPS: 500 # your max episode steps\n\n### MODEL（YOUR TRAINING MODEL CONFIG）\nMODEL_NAME: \"DQN\" # \"PPO\" or \"DQN\"\nMODEL_POLICY: \"MlpPolicy\" # \"MlpPolicy\" or \"CnnPolicy\"\nMODEL_KWARGS: \n \"policy_kwargs\": \n net_arch: [256, 256]\n \"learning_rate\": 0.0005 # 5e-4\n \"buffer_size\": 15000\n \"learning_starts\": 64 # 通常应为 batch_size 的 1-2 倍\n \"batch_size\": 32\n \"gamma\": 0.99\n \"tau\": 1.0\n \"train_freq\": 1\n \"gradient_steps\": 1\n \"target_update_interval\": 50\n \"exploration_fraction\": 0.1 # 探索衰减比例，前10%的训练时间内从ε_initial衰减到ε_final\n \"exploration_initial_eps\": 1.0 # 初始探索率，开始时100%随机选择动作\n \"exploration_final_eps\": 0.02 # 最终探索率，训练后期仍保持2%的随机性\n \"max_grad_norm\": 0.5 # 梯度裁剪(?>0.5->?==0.5)\n \"verbose\": 1\n \"tensorboard_log\": \"../reproduce1/highway_dqn/\"\n\n\nreward_threshold: 9000 # your reward threshold\ncycle_number: 2 # your cycle number\nevery_cycle_timesteps: 1000 # total_timesteps/cycle number\neval_freq: 1000 # your eval freq and save model frequency \nn_eval_episodes: 5 # your n eval episodes 5\n\ntensorboard_log_path: \"../reproduce1/highway_dqn/\" # your model save path\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n### WASTE SPACE\nLLM_TOOLS: ## 定义LLM用到 读取文件的工具\n - \"type\": \"function\"\n \"function\": \n \"name\": \"read_file\"\n \"description\": \"Read the contents of a local file\"\n \"parameters\": \n \"type\": \"object\"\n \"properties\": \n \"path\": \n \"type\": \"string\"\n \"enum\": # 只允许调用这些文件\n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\"\n \"required\": \n - \"path\"\n# your path to specific env files\nENV_FILE_PATH: \n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\""}, "60a9dbff19e4ccc61f4fa54900c272d5a59bcc54": {"t": 1785761663.5995545, "n": 3387, "h": "bbf98f76dc2594be", "c": "### 如环境变动要自己去环境仓库 看需要导入哪些相关文件\n\n# YOUR LLMs API CONFIG\n### DEEPSEEK\nDEEPSEEK_KEY: \"sk-1495882192234e668f470fba3e39f9ce\"\nDEEPSEEK_MODEL: \"deepseek-reasoner\"\nDEEPSEEK_BASE_URL: \"https://api.deepseek.com\"\nDEEPSEEK_MAX_TOKENS: 64000 # 8192 最大输出tokens\nDEEPSEEK_TEMPERATURE: 0.3 # reasoner用不了温度\n### qwen\nQWEN_KEY: \"sk-1c71165453eb4ad192753babda9bffe0\"\nQWEN_MODEL: \"qwen3-coder-plus\"\nQWEN_BASE_URL: \"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nQWEN_MAX_INPUT_TOKENS: 997,952 # 最大输入tokens\nQWEN_MAX_TOKENS: 65,536 # 最大输出tokens\nQWEN_ENABLE_THINKING: True # 是否开启思考模式\n\n### ENV FILE ADDRESS\nENV_FILE_ADDRESS: \"../HighwayEnv-RL_env/highway_env\" # your path to the all env file\nENV_ID: \"highway-v0\"\nMAX_EPISODE_STEPS: 1000 # your max episode steps\n\n### MODEL（YOUR TRAINING MODEL CONFIG）\nMODEL_NAME: \"DQN\" # \"PPO\" or \"DQN\"\nMODEL_POLICY: \"MlpPolicy\" # \"MlpPolicy\" or \"CnnPolicy\"\nMODEL_KWARGS: \n \"policy_kwargs\": \n net_arch: [256, 256]\n \"learning_rate\": 0.0005 # 5e-4\n \"buffer_size\": 15000\n \"learning_starts\": 200\n \"batch_size\": 32\n \"gamma\": 0.8\n \"tau\": 1.0\n \"train_freq\": 1\n \"gradient_steps\": 1\n \"target_update_interval\": 50\n \"exploration_fraction\": 0.1 # 探索衰减比例，前10%的训练时间内从ε_initial衰减到ε_final\n \"exploration_initial_eps\": 1.0 # 初始探索率，开始时100%随机选择动作\n \"exploration_final_eps\": 0.02 # 最终探索率，训练后期仍保持2%的随机性\n \"max_grad_norm\": 0.5 # 梯度裁剪(?>0.5->?==0.5)\n \"verbose\": 1\n \"tensorboard_log\": \"../reproduce1/highway_dqn/\"\n\n\nreward_threshold: 9000 # your reward threshold\ncycle_number: 10 # your cycle number\nevery_cycle_timesteps: 200000 # total_timesteps/cycle number\neval_freq: 10000 # your eval freq and save model frequency \nn_eval_episodes: 5 # your n eval episodes 5\n\ntensorboard_log_path: \"../reproduce1/highway_dqn/\" # your model save path\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n### WASTE SPACE\nLLM_TOOLS: ## 定义LLM用到 读取文件的工具\n - \"type\": \"function\"\n \"function\": \n \"name\": \"read_file\"\n \"description\": \"Read the contents of a local file\"\n \"parameters\": \n \"type\": \"object\"\n \"properties\": \n \"path\": \n \"type\": \"string\"\n \"enum\": # 只允许调用这些文件\n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\"\n \"required\": \n - \"path\"\n# your path to specific env files\nENV_FILE_PATH: \n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\""}, "751e370bf2e793505f011f111bcabb15fff89466": {"t": 1785761666.8294587, "n": 3316, "h": "ed236c4c8b2bfaa0", "c": "### 如环境变动要自己去环境仓库 看需要导入哪些相关文件\n\n# YOUR LLMs API CONFIG\n### DEEPSEEK\nDEEPSEEK_KEY: \"sk-1495882192234e668f470fba3e39f9ce\"\nDEEPSEEK_MODEL: \"deepseek-reasoner\"\nDEEPSEEK_BASE_URL: \"https://api.deepseek.com\"\nDEEPSEEK_MAX_TOKENS: 64000 # 8192 最大输出tokens\nDEEPSEEK_TEMPERATURE: 0.3 # reasoner用不了温度\n### qwen\nQWEN_KEY: \"sk-1c71165453eb4ad192753babda9bffe0\"\nQWEN_MODEL: \"qwen3-coder-plus\"\nQWEN_BASE_URL: \"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nQWEN_MAX_INPUT_TOKENS: 997,952 # 最大输入tokens\nQWEN_MAX_TOKENS: 65,536 # 最大输出tokens\nQWEN_ENABLE_THINKING: True # 是否开启思考模式\n\n### ENV FILE ADDRESS\nENV_FILE_ADDRESS: \"../HighwayEnv-RL_env/highway_env\" # your path to the all env file\n\n\n### MODEL（YOUR TRAINING MODEL CONFIG）\nMODEL_NAME: \"DQN\" # \"PPO\" or \"DQN\"\nMODEL_POLICY: \"MlpPolicy\" # \"MlpPolicy\" or \"CnnPolicy\"\nMODEL_KWARGS: \n \"policy_kwargs\": \n net_arch: [256, 256]\n \"learning_rate\": 0.0005 # 5e-4\n \"buffer_size\": 15000\n \"learning_starts\": 200\n \"batch_size\": 32\n \"gamma\": 0.8\n \"tau\": 1.0\n \"train_freq\": 1\n \"gradient_steps\": 1\n \"target_update_interval\": 50\n \"exploration_fraction\": 0.1 # 探索衰减比例，前10%的训练时间内从ε_initial衰减到ε_final\n \"exploration_initial_eps\": 1.0 # 初始探索率，开始时100%随机选择动作\n \"exploration_final_eps\": 0.02 # 最终探索率，训练后期仍保持2%的随机性\n \"max_grad_norm\": 0.5 # 梯度裁剪(?>0.5->?==0.5)\n \"verbose\": 1\n \"tensorboard_log\": \"../reproduce1/highway_dqn/\"\n\n\nreward_threshold: 9000 # your reward threshold\ncycle_number: 10 # your cycle number\nevery_cycle_timesteps: 200000 # total_timesteps/cycle number\neval_freq: 10000 # your eval freq and save model frequency \nn_eval_episodes: 5 # your n eval episodes 5\n\ntensorboard_log_path: \"../reproduce1/highway_dqn/\" # your model save path\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n### WASTE SPACE\nLLM_TOOLS: ## 定义LLM用到 读取文件的工具\n - \"type\": \"function\"\n \"function\": \n \"name\": \"read_file\"\n \"description\": \"Read the contents of a local file\"\n \"parameters\": \n \"type\": \"object\"\n \"properties\": \n \"path\": \n \"type\": \"string\"\n \"enum\": # 只允许调用这些文件\n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\"\n \"required\": \n - \"path\"\n# your path to specific env files\nENV_FILE_PATH: \n - \"../HighwayEnv-RL_env/highway_env/envs/highway_env.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/abstract.py\"\n - \"../HighwayEnv-RL_env/highway_env/road/road.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/controller.py\"\n - \"../HighwayEnv-RL_env/highway_env/vehicle/kinematics.py\"\n - \"../HighwayEnv-RL_env/highway_env/envs/common/action.py\"\n - \"../HighwayEnv-RL_env/highway_env/utils.py\""}}}} |