Feature: 提供了主题识别的配置选项，现在可以自己选模型了

Feature: 提供了主题识别的配置选项，现在可以自己选模型了Merge pull request #49 from tcmofashi/feature

config/bot_config_template.toml

@@ -55,6 +55,8 @@ ban_user_id = []  #禁止回复消息的QQ号
 #base_url = "DEEP_SEEK_BASE_URL"
 #key = "DEEP_SEEK_KEY"
 
+#下面的模型若使用硅基流动则不需要更改，使用ds官方则改成.env.prod自定义的宏，使用自定义模型则选择定位相似的模型自己填写
+
 [model.llm_reasoning] #R1
 name = "Pro/deepseek-ai/DeepSeek-R1"
 base_url = "SILICONFLOW_BASE_URL"
@@ -84,3 +86,12 @@ key = "SILICONFLOW_KEY"
 name = "BAAI/bge-m3"
 base_url = "SILICONFLOW_BASE_URL"
 key = "SILICONFLOW_KEY"
+
+# 主题提取，jieba和snownlp不用api，llm需要api
+[topic]
+topic_extract='snownlp' # 只支持jieba,snownlp,llm三种选项
+
+[topic.llm_topic]
+name = "Pro/deepseek-ai/DeepSeek-V3"
+base_url = "SILICONFLOW_BASE_URL"
+key = "SILICONFLOW_KEY"

src/plugins/chat/bot.py

@@ -67,13 +67,18 @@ class ChatBot:
         
         current_time = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(message.time))
 
+
+        identifier=topic_identifier.identify_topic()
+        if global_config.topic_extract=='llm':
+            topic=await identifier(message.processed_plain_text)
+        else:
+            topic=identifier(message.detailed_plain_text)
+
         # topic1 = topic_identifier.identify_topic_jieba(message.processed_plain_text)
         # topic2 = await topic_identifier.identify_topic_llm(message.processed_plain_text)
-        topic3 = topic_identifier.identify_topic_snownlp(message.processed_plain_text)
-        # print(f"\033[1;32m[主题识别]\033[0m 使用jieba主题: {topic1}")
-        # print(f"\033[1;32m[主题识别]\033[0m 使用llm主题: {topic2}")
-        print(f"\033[1;32m[主题识别]\033[0m 使用snownlp主题: {topic3}")
-        topic = topic3
+        # topic3 = topic_identifier.identify_topic_snownlp(message.processed_plain_text)
+        print(f"\033[1;32m[主题识别]\033[0m 使用{global_config.topic_extract}主题: {topic}")
+
         
         all_num = 0
         interested_num = 0

src/plugins/chat/config.py

@@ -42,6 +42,10 @@ class BotConfig:
     embedding: Dict[str, str] = field(default_factory=lambda: {})
     vlm: Dict[str, str] = field(default_factory=lambda: {})
 
+    # 主题提取配置
+    topic_extract: str = 'snownlp' # 只支持jieba,snownlp,llm
+    llm_topic_extract: Dict[str, str] = field(default_factory=lambda: {})
+    
     API_USING: str = "siliconflow"  # 使用的API
     API_PAID: bool = False  # 是否使用付费API
     MODEL_R1_PROBABILITY: float = 0.8  # R1模型概率
@@ -81,8 +85,10 @@ class BotConfig:
                 personality_config=toml_dict['personality']
                 personality=personality_config.get('prompt_personality')
                 if len(personality) >= 2:
-                    config.PROMPT_PERSONALITY=personality_config.get('prompt_personality')
-                config.PROMPT_SCHEDULE_GEN=personality_config.get('prompt_schedule')
+                    print(f"载入自定义人格:{personality}")
+                    config.PROMPT_PERSONALITY=personality_config.get('prompt_personality',config.PROMPT_PERSONALITY)
+                print(f"载入自定义日程prompt:{personality_config.get('prompt_schedule',config.PROMPT_SCHEDULE_GEN)}")
+                config.PROMPT_SCHEDULE_GEN=personality_config.get('prompt_schedule',config.PROMPT_SCHEDULE_GEN)
 
             if "emoji" in toml_dict:
                 emoji_config = toml_dict["emoji"]
@@ -120,6 +126,7 @@ class BotConfig:
                 
                 if "llm_normal" in model_config:
                     config.llm_normal = model_config["llm_normal"]
+                    config.llm_topic_extract = config.llm_normal
                 
                 if "llm_normal_minor" in model_config:
                     config.llm_normal_minor = model_config["llm_normal_minor"]
@@ -130,6 +137,15 @@ class BotConfig:
                 if "embedding" in model_config:
                     config.embedding = model_config["embedding"]
                 
+            if 'topic' in toml_dict:
+                topic_config=toml_dict['topic']
+                if 'topic_extract' in topic_config:
+                    config.topic_extract=topic_config.get('topic_extract',config.topic_extract)
+                    print(f"载入自定义主题提取为{config.topic_extract}")
+                if config.topic_extract=='llm' and 'llm_topic' in topic_config:
+                    config.llm_topic_extract=topic_config['llm_topic']
+                    print(f"载入自定义主题提取模型为{config.llm_topic_extract['name']}")
+                
             # 消息配置
             if "message" in toml_dict:
                 msg_config = toml_dict["message"]

src/plugins/chat/topic_identifier.py

@@ -12,7 +12,19 @@ config = driver.config
 
 class TopicIdentifier:
     def __init__(self):
-        self.llm_client = LLM_request(model=global_config.llm_normal)
+        self.llm_client = LLM_request(model=global_config.llm_topic_extract)
+        self.select=global_config.topic_extract
+
+    def identify_topic(self):
+        if self.select=='jieba':
+            return self.identify_topic_jieba
+        elif self.select=='snownlp':
+            return self.identify_topic_snownlp
+        elif self.select=='llm':
+            return self.identify_topic_llm
+        else:
+            return self.identify_topic_snownlp
+
         
     async def identify_topic_llm(self, text: str) -> Optional[List[str]]:
         """识别消息主题，返回主题列表"""