增加了多阶段模型参数的单独读取

config/ai_model.json

@@ -2,9 +2,9 @@
     "model": "qwen-plus",
     "api_url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
     "api_key": "sk-bd5ee62703bc41fc9b8a55d748dc1eb8",
-    "temperature": 0.8,
+    "temperature": 0.3,
-    "top_p": 0.5,
+    "top_p": 0.4,
-    "presence_penalty": 1.1,
+    "presence_penalty": 1.2,
     "timeout": 120,
     "max_retries": 3
 } 

config/content_gen.json

@@ -3,5 +3,15 @@
     "content_user_prompt": "resource/prompt/generateContent/user.txt",
     "judger_system_prompt": "resource/prompt/judgeContent/system.txt",
     "judger_user_prompt": "resource/prompt/judgeContent/user.txt",
-    "enable_content_judge": true
+    "enable_content_judge": true,
+    "model": {
+        "temperature": 0.3,
+        "top_p": 0.5,
+        "presence_penalty": 1.2
+    },
+    "judger_model": {
+        "temperature": 0.2,
+        "top_p": 0.3,
+        "presence_penalty": 0.8
+    }
 } 

config/resource.json

@@ -16,7 +16,7 @@
         "refer_list": [
             { "path": "resource/prompt/Refer/2025各月节日宣传节点时间表.md", "sampling_rate": 1, "step": "topic" },
             { "path": "resource/prompt/Refer/标题参考格式.json", "sampling_rate": 0.25, "step": "content" },
-            { "path": "resource/prompt/Refer/正文范文参考.json", "sampling_rate": 0.25, "step": "content" }
+            { "path": "resource/prompt/Refer/正文范文参考.json", "sampling_rate": 0.5, "step": "content" }
         ]
     },
     "object": {

config/topic_gen.json

@@ -1,7 +1,11 @@
 {
     "topic_system_prompt": "resource/prompt/generateTopics/system.txt",
     "topic_user_prompt": "resource/prompt/generateTopics/user.txt",
-    "model": {},
+    "model": {
+        "temperature": 0.2,
+        "top_p": 0.3,
+        "presence_penalty": 1.5
+    },
     "topic": {
         "date": "2024-07-20",
         "num": 5,

core/ai/ai_agent.py

@@ -45,7 +45,9 @@ class AIAgent:
         #     self.tokenizer = tiktoken.get_encoding("cl100k_base")
 
     async def generate_text(
-        self, system_prompt: str, user_prompt: str, use_stream: bool = False
+        self, system_prompt: str, user_prompt: str, use_stream: bool = False,
+        temperature: Optional[float] = None, top_p: Optional[float] = None, 
+        presence_penalty: Optional[float] = None, stage: str = ""
     ) -> Tuple[str, int, int, float]:
         """
         生成文本 (支持流式和非流式)
@@ -54,6 +56,10 @@ class AIAgent:
             system_prompt: 系统提示
             user_prompt: 用户提示
             use_stream: 是否流式返回
+            temperature: 温度参数，控制随机性
+            top_p: Top-p采样参数
+            presence_penalty: 存在惩罚参数
+            stage: 当前所处阶段，用于日志记录
 
         Returns:
             一个元组 (generated_text, input_tokens, output_tokens, time_cost)
@@ -63,8 +69,17 @@ class AIAgent:
             {"role": "user", "content": user_prompt}
         ]
         
+        # 使用传入的参数或默认配置
+        temp = temperature if temperature is not None else self.config.temperature
+        tp = top_p if top_p is not None else self.config.top_p
+        pp = presence_penalty if presence_penalty is not None else self.config.presence_penalty
+        
+        # 记录使用的模型参数
+        stage_info = f"[{stage}]" if stage else ""
+        logger.info(f"{stage_info} 使用模型参数: temperature={temp:.2f}, top_p={tp:.2f}, presence_penalty={pp:.2f}")
+        
         input_tokens = self.count_tokens(system_prompt + user_prompt)
-        logger.info(f"开始生成任务... 输入token数: {input_tokens}")
+        logger.info(f"{stage_info} 开始生成任务... 输入token数: {input_tokens}")
 
         last_exception = None
         backoff_time = 1.0  # Start with 1 second
@@ -75,9 +90,9 @@ class AIAgent:
                 response = await self.client.chat.completions.create(
                     model=self.config.model,
                     messages=messages,
-                    temperature=self.config.temperature,
+                    temperature=temp,
-                    top_p=self.config.top_p,
+                    top_p=tp,
-                    presence_penalty=self.config.presence_penalty,
+                    presence_penalty=pp,
                     stream=use_stream
                 )
                 
@@ -90,26 +105,27 @@ class AIAgent:
                     # 简化处理：流式模式下，我们返回拼接后的完整文本
                     full_text = "".join([chunk for chunk in self._process_stream(response)])
                     output_tokens = self.count_tokens(full_text)
+                    logger.info(f"{stage_info} 任务完成，耗时 {time_cost:.2f} 秒. 输出token数: {output_tokens}")
                     return full_text, input_tokens, output_tokens, time_cost
                 else:
                     output_text = response.choices[0].message.content.strip()
                     output_tokens = self.count_tokens(output_text)
-                    logger.info(f"任务完成，耗时 {time_cost:.2f} 秒. 输出token数: {output_tokens}")
+                    logger.info(f"{stage_info} 任务完成，耗时 {time_cost:.2f} 秒. 输出token数: {output_tokens}")
                     return output_text, input_tokens, output_tokens, time_cost
 
             except (APITimeoutError, APIConnectionError) as e:
                 last_exception = RetryableError(f"AI模型连接或超时错误: {e}")
-                logger.warning(f"尝试 {attempt + 1}/{self.config.max_retries} 失败: {last_exception}. "
+                logger.warning(f"{stage_info} 尝试 {attempt + 1}/{self.config.max_retries} 失败: {last_exception}. "
                                f"将在 {backoff_time:.1f} 秒后重试...")
                 time.sleep(backoff_time)
                 backoff_time *= 2  # Exponential backoff
             except (RateLimitError, APIStatusError) as e:
                 last_exception = NonRetryableError(f"AI模型API错误 (不可重试): {e}")
-                logger.error(f"发生不可重试的API错误: {last_exception}")
+                logger.error(f"{stage_info} 发生不可重试的API错误: {last_exception}")
                 break  # Do not retry on these errors
             except Exception as e:
                 last_exception = AIModelError(f"调用AI模型时发生未知错误: {e}")
-                logger.error(f"发生未知错误: {last_exception}\n{traceback.format_exc()}")
+                logger.error(f"{stage_info} 发生未知错误: {last_exception}\n{traceback.format_exc()}")
                 break
 
         raise AIModelError(f"AI模型调用在 {self.config.max_retries} 次重试后失败") from last_exception

utils/tweet/content_generator.py

@@ -10,7 +10,7 @@ import json
 from typing import Dict, Any
 
 from core.ai import AIAgent
-from core.config import ConfigManager, GenerateTopicConfig
+from core.config import ConfigManager, GenerateTopicConfig, GenerateContentConfig
 from utils.prompts import ContentPromptBuilder
 from utils.file_io import OutputManager
 
@@ -22,7 +22,9 @@ class ContentGenerator:
 
     def __init__(self, ai_agent: AIAgent, config_manager: ConfigManager, output_manager: OutputManager):
         self.ai_agent = ai_agent
-        self.config: GenerateTopicConfig = config_manager.get_config('topic_gen', GenerateTopicConfig)
+        self.config_manager = config_manager
+        self.topic_config = config_manager.get_config('topic_gen', GenerateTopicConfig)
+        self.content_config = config_manager.get_config('content_gen', GenerateContentConfig)
         self.output_manager = output_manager
         self.prompt_builder = ContentPromptBuilder(config_manager)
 
@@ -49,12 +51,25 @@ class ContentGenerator:
         self.output_manager.save_text(system_prompt, "content_system_prompt.txt", subdir=output_dir.name)
         self.output_manager.save_text(user_prompt, "content_user_prompt.txt", subdir=output_dir.name)
 
+        # 获取模型参数
+        model_params = {}
+        if hasattr(self.content_config, 'model') and isinstance(self.content_config.model, dict):
+            model_params = {
+                'temperature': self.content_config.model.get('temperature'),
+                'top_p': self.content_config.model.get('top_p'),
+                'presence_penalty': self.content_config.model.get('presence_penalty')
+            }
+            # 移除None值
+            model_params = {k: v for k, v in model_params.items() if v is not None}
+
         # 2. 调用AI
         try:
             raw_result, _, _, _ = await self.ai_agent.generate_text(
                 system_prompt=system_prompt,
                 user_prompt=user_prompt,
-                use_stream=False
+                use_stream=False,
+                stage="内容生成",
+                **model_params
             )
             self.output_manager.save_text(raw_result, "content_raw_response.txt", subdir=output_dir.name)
         except Exception as e:

utils/tweet/content_judger.py

@@ -11,7 +11,7 @@ from json_repair import loads as json_repair_loads
 from typing import Dict, Any
 
 from core.ai import AIAgent
-from core.config import ConfigManager, GenerateTopicConfig
+from core.config import ConfigManager, GenerateTopicConfig, GenerateContentConfig
 from utils.prompts import JudgerPromptBuilder
 
 logger = logging.getLogger(__name__)
@@ -30,7 +30,9 @@ class ContentJudger:
             output_manager: 输出管理器，用于保存提示词和响应
         """
         self.ai_agent = ai_agent
-        self.config: GenerateTopicConfig = config_manager.get_config('topic_gen', GenerateTopicConfig)
+        self.config_manager = config_manager
+        self.topic_config = config_manager.get_config('topic_gen', GenerateTopicConfig)
+        self.content_config = config_manager.get_config('content_gen', GenerateContentConfig)
         self.prompt_builder = JudgerPromptBuilder(config_manager)
         self.output_manager = output_manager
 
@@ -63,12 +65,25 @@ class ContentJudger:
             self.output_manager.save_text(system_prompt, f"{topic_dir}/judger_system_prompt.txt")
             self.output_manager.save_text(user_prompt, f"{topic_dir}/judger_user_prompt.txt")
 
+        # 获取模型参数
+        model_params = {}
+        if hasattr(self.content_config, 'judger_model') and isinstance(self.content_config.judger_model, dict):
+            model_params = {
+                'temperature': self.content_config.judger_model.get('temperature'),
+                'top_p': self.content_config.judger_model.get('top_p'),
+                'presence_penalty': self.content_config.judger_model.get('presence_penalty')
+            }
+            # 移除None值
+            model_params = {k: v for k, v in model_params.items() if v is not None}
+
         # 2. 调用AI进行审核
         try:
             raw_result, _, _, _ = await self.ai_agent.generate_text(
                 system_prompt=system_prompt,
                 user_prompt=user_prompt,
-                use_stream=False
+                use_stream=False,
+                stage="内容审核",
+                **model_params
             )
             
             # 保存原始响应

utils/tweet/topic_generator.py

@@ -55,12 +55,25 @@ class TopicGenerator:
         self.output_manager.save_text(system_prompt, "topic_system_prompt.txt")
         self.output_manager.save_text(user_prompt, "topic_user_prompt.txt")
         
+        # 获取模型参数
+        model_params = {}
+        if hasattr(self.config, 'model') and isinstance(self.config.model, dict):
+            model_params = {
+                'temperature': self.config.model.get('temperature'),
+                'top_p': self.config.model.get('top_p'),
+                'presence_penalty': self.config.model.get('presence_penalty')
+            }
+            # 移除None值
+            model_params = {k: v for k, v in model_params.items() if v is not None}
+        
         # 2. 调用AI生成
         try:
             raw_result, _, _, _ = await self.ai_agent.generate_text(
                 system_prompt=system_prompt,
                 user_prompt=user_prompt,
-                use_stream=False # 选题生成通常不需要流式输出
+                use_stream=False, # 选题生成通常不需要流式输出
+                stage="选题生成",
+                **model_params
             )
             self.output_manager.save_text(raw_result, "topics_raw_response.txt")
         except Exception as e: