From 1f6ba5fd092ed49de7110dad569700597e4b95f6 Mon Sep 17 00:00:00 2001 From: zcr Date: Tue, 23 Dec 2025 16:38:40 +0800 Subject: [PATCH] =?UTF-8?q?feat=EF=BC=9Aupdate=20chatbot=20sys=20prompt?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .gitea/workflows/prod_build_manual.yaml | 4 +- app/server/ChatbotAgent/chatbot_server.py | 55 ++++++++++++++++------- app/server/ChatbotAgent/core/prompt.py | 44 +++++++++++++++--- 3 files changed, 79 insertions(+), 24 deletions(-) diff --git a/.gitea/workflows/prod_build_manual.yaml b/.gitea/workflows/prod_build_manual.yaml index df673f7..a9f6469 100644 --- a/.gitea/workflows/prod_build_manual.yaml +++ b/.gitea/workflows/prod_build_manual.yaml @@ -36,5 +36,5 @@ jobs: docker-compose down 2>&1 docker-compose up -d --build --remove-orphans 2>&1 - - docker image prune -f 2>&1 \ No newline at end of file + +# docker image prune -f 2>&1 \ No newline at end of file diff --git a/app/server/ChatbotAgent/chatbot_server.py b/app/server/ChatbotAgent/chatbot_server.py index 5cd5005..971af7a 100644 --- a/app/server/ChatbotAgent/chatbot_server.py +++ b/app/server/ChatbotAgent/chatbot_server.py @@ -1,9 +1,13 @@ import logging +import sys + import litserve as ls from typing import AsyncGenerator from google import genai from pydantic import BaseModel +from sympy.core.evalf import rnd + from app.config import settings from google.genai import types @@ -95,42 +99,61 @@ class LCChatBot(ls.LitAPI): if __name__ == "__main__": + sys.stdout = open('permanent.log', 'w', encoding='utf-8') + import asyncio - async def run_simple_test(): + + + async def run_simple_test(text): """ 一个简单的异步测试用例，用于测试 LCChatBot 的流式输出。 """ print("\n" + "=" * 50) - print("--- 🔬 开始 LCChatBot 简单流式测试 ---") - + # print("--- 🔬 开始 LCChatBot 简单流式测试 ---") + # 1. 初始化 LitAPI 和其依赖 chatbot_api = LCChatBot() - chatbot_api.setup(device="cpu") - print("✅ Setup complete. Mock services initialized.") + chatbot_api.setup(device="cpu") + # print("✅ Setup complete. Mock services initialized.") # 2. 构造请求数据 request_data = PredictRequest( user_id="simple_user", session_id="simple_session", - user_message="I want an outfit. I am going to a evening party with friends. Suggest something stylish yet comfortable.", + user_message=text, gender="female" ) chatbot_api.redis.clear_history(request_data.session_id) - print(f"-> 正在发送查询: {request_data.user_message}") - + print(f"user: \n {request_data.user_message}") + # 3. 调用 predict 方法并处理流 response_generator = chatbot_api.predict(request_data) - - print("\n<- 接收流式响应:") + + print("agent:") # 4. 异步迭代生成器，实时打印输出 async for chunk in response_generator: print(chunk, end="", flush=True) - print("\n" + "=" * 50) - # 启动异步事件循环 - try: - asyncio.run(run_simple_test()) - except Exception as e: - print(f"\n发生致命错误: {e}") \ No newline at end of file + + text_list = [ + '我要去参加好朋友的婚礼，你能帮我挑一套衣服吗？', + 'I need something to wear for a big presentation at work tomorrow. I want to look powerful but still approachable.', + 'Who do you think is the best world leader right now?', + 'I’m going on a trip to Paris next week and need some outfits.', + 'Help me find a cool outfit for a rock concert. I hate wearing dresses.', + 'I want to look very cool, 或者是那种很有个性的风格 for a gallery opening.', + "I'm going to a gala. Please list 5 different dress styles for me and use bold text for the names.", + "I'm feeling really sad today and just want an outfit that matches my mood." + ] + for text in text_list: + asyncio.run(run_simple_test(text)) + # print("\n" + "=" * 50) + # # 启动异步事件循环 + # try: + # asyncio.run(run_simple_test()) + # except Exception as e: + # print(f"\n发生致命错误: {e}") + # + sys.stdout.close() diff --git a/app/server/ChatbotAgent/core/prompt.py b/app/server/ChatbotAgent/core/prompt.py index 79dbf34..c5633ae 100644 --- a/app/server/ChatbotAgent/core/prompt.py +++ b/app/server/ChatbotAgent/core/prompt.py @@ -1,4 +1,4 @@ -BASIC_PROMPT = """You are a professional, friendly, and insightful AI {gender}'s styling assistant. +BASIC_PROMPT_OLD = """You are a professional, friendly, and insightful AI {gender}'s styling assistant. Your primary mission is to engage in a multi-turn conversation with the user to fully understand their dressing intent. You must adopt a professional yet approachable tone. @@ -21,6 +21,39 @@ Example Follow-up (mimicking a conversational flow): User: I want a chic outfit for dinner. Your Response: Hey there! A chic dinner outfit, I love that! To give you the perfect recommendations, tell me: is this a romantic date, business dinner, or celebration with friends? And what's your go-to style vibe: classic elegance or something with more edge?""" +BASIC_PROMPT = """ +You are a professional, friendly, and insightful AI {gender}'s styling assistant. You are smart, young, and enthusiastic, turning styling into an exciting experience. Your tone is warm, confident, composed, and genuinely curious about the user's context. + +CONVERSATION GOALS: + 1. **Occasion:** Determine the specific event (e.g., romantic dinner, summer wedding). + 2. **Style:** Pinpoint the desired aesthetic (e.g., minimalist, bohemian). + 3. **Vibe/Details:** Gather mood or constraints (e.g., comfort, specific colors). + 4. **Item Preference:** Identify preferences for silhouettes or specific items. + +PERSONALITY AND COMMUNICATION RULES: + - Always reply in English only. If the user speaks another language, state in English that you cannot speak that language and steer back to styling. + - Keep responses extremely SHORT (maximum 2 sentences). + - Use clear, modern, and vivid language. Avoid clichés or corporate jargon. + - Be positive and encouraging, never patronizing. + - When uncertain, state assumptions and invite correction. + - Gather sufficient intent before providing recommendations. Ask only 1 to 2 essential questions per turn. + +IMPORTANT GUARDRAILS: + - ONLY help with styling ideas and outfit generation. + - If asked about politics, controversial topics, or anything inappropriate, respond with a witty deflection and steer back to styling. + +OUTPUT FORMAT INSTRUCTION: +- **DO NOT** use any Markdown formatting whatsoever (no asterisks, bold text, lists, or code blocks). +- **ONLY** output the plain text response spoken by the AI Assistant. + +EXAMPLE DIALOGUES: + User: 我想找件衣服参加婚礼。 Response: I am so sorry, but I can only speak English! I would love to help you find the perfect wedding guest look if you can tell me the dress code or the venue vibe. + + User: I need a chic outfit for dinner. Response: That sounds like a fabulous evening! Is this a romantic date or a casual night out with friends, and do you prefer a sleek dress or tailored separates? + + User: What do you think about the current election? Response: I prefer to stick to the polls of fashion rather than politics! Tell me, what kind of style aesthetic are we aiming for today? +""" + SUMMARY_PROMPT = """ You are an expert fashion request analyzer. Analyze the conversation history provided by the user. Your task is to: @@ -31,8 +64,8 @@ Your task is to: Extract this information accurately from the chat history. """ - from app.taxonomy import FASHION_TAXONOMY, IGNORE_SUBCATEGORY, ALL_SUBCATEGORY_LIST + core_outfit_template = f""" You are a professional fashion stylist Agent, specialized in creating complete, tailored outfits for {{gender}}. Your current task is to recommend items for the **{{current_category}}** stage, strictly **mimicking the style and preference** specified in the following Stylist Guide. @@ -96,7 +129,6 @@ You must strictly use the **JSON format** for your output, as follows: Now, please start building an outfit (with strictly unique categories for all items) and output the JSON for the first item. """ - accessories_template = f""" You are a professional fashion stylist Agent, specialized in creating complete, tailored outfits for {{gender}}. Your current task is to finalize the look by recommending accessories for the **{{current_category}}** stage, strictly **mimicking the style and preference** specified in the following Accessories Guide. @@ -199,7 +231,7 @@ def build_iterative_schema(current_category): "enum": FASHION_TAXONOMY[current_category] }, "description": { - "type": "string", + "type": "string", "description": "an **extremely detailed and precise** description of the item. This description is used for **high-accuracy vector search** in the database. It should include Color, Fit/Silhouette, Material/Detail, Role in the Outfit." }, "reason": {"type": "string", "description": "The reason for the current action. Required if action is 'stop' (to summarize the final outfit)."} @@ -209,8 +241,8 @@ def build_iterative_schema(current_category): return schema -def build_batch_schema(specified_category: str=""): - assert(specified_category in FASHION_TAXONOMY.keys() or specified_category == "") +def build_batch_schema(specified_category: str = ""): + assert (specified_category in FASHION_TAXONOMY.keys() or specified_category == "") if not specified_category: category_range_desc = "the complete final outfit (including all categories)" subcategory_list = ALL_SUBCATEGORY_LIST