323 lines
13 KiB
Python
323 lines
13 KiB
Python
# 基于知识库提问的通用模板，
|
||
# assistant_id
|
||
import json
|
||
import os
|
||
import re
|
||
import queue
|
||
import concurrent.futures
|
||
import time
|
||
import requests
|
||
from dashscope import Assistants, Messages, Runs, Threads
|
||
from llama_index.indices.managed.dashscope import DashScopeCloudRetriever
|
||
from flask_app.main.通义千问long import qianwen_long, upload_file
|
||
prompt = """
|
||
# 角色
|
||
你是一个文档处理专家，专门负责理解和操作基于特定内容的文档任务，这包括解析、总结、搜索或生成与给定文档相关的各类信息。
|
||
|
||
## 技能
|
||
### 技能 1：文档解析与摘要
|
||
- 深入理解并分析${documents}的内容，提取关键信息。
|
||
- 根据需求生成简洁明了的摘要，保持原文核心意义不变。
|
||
|
||
### 技能 2：信息检索与关联
|
||
- 在${documents}中高效检索特定信息或关键词。
|
||
- 能够识别并链接到文档内部或外部的相关内容，增强信息的连贯性和深度。
|
||
|
||
## 限制
|
||
- 所有操作均需基于${documents}的内容，不可超出此范围创造信息。
|
||
- 在处理敏感或机密信息时，需遵守严格的隐私和安全规定。
|
||
- 确保所有生成或改编的内容逻辑连贯，无误导性信息。
|
||
|
||
请注意，上述技能执行时将直接利用并参考${documents}的具体内容，以确保所有产出紧密相关且高质量。
|
||
"""
|
||
prom = '请记住以下材料，他们对回答问题有帮助，请你简洁准确地给出回答，不要给出无关内容。${documents}'
|
||
|
||
def read_questions_from_file(file_path):
|
||
questions = []
|
||
current_question = ""
|
||
current_number = 0
|
||
|
||
with open(file_path, 'r', encoding='utf-8') as file:
|
||
for line in file:
|
||
line = line.strip()
|
||
if not line: # 跳过空行
|
||
continue
|
||
|
||
# 检查是否是新的问题编号
|
||
match = re.match(r'^(\d+)\.', line)
|
||
if match:
|
||
# 如果有之前的问题，保存它
|
||
if current_question:
|
||
questions.append(current_question.strip())
|
||
|
||
# 开始新的问题
|
||
current_number = int(match.group(1))
|
||
current_question = line.split('.', 1)[1].strip() + "\n"
|
||
else:
|
||
# 继续添加到当前问题
|
||
current_question += line + "\n"
|
||
|
||
# 添加最后一个问题
|
||
if current_question:
|
||
questions.append(current_question.strip())
|
||
|
||
return questions
|
||
|
||
|
||
#正文和文档名之间的内容
|
||
|
||
def send_message(assistant, message='百炼是什么？'):
|
||
ans = []
|
||
print(f"Query: {message}")
|
||
|
||
# create thread.
|
||
thread = Threads.create()
|
||
print(thread)
|
||
|
||
# create a message.
|
||
message = Messages.create(thread.id, content=message)
|
||
# create run
|
||
|
||
run = Runs.create(thread.id, assistant_id=assistant.id)
|
||
# print(run)
|
||
|
||
# wait for run completed or requires_action
|
||
run_status = Runs.wait(run.id, thread_id=thread.id)
|
||
# print(run_status)
|
||
# get the thread messages.
|
||
msgs = Messages.list(thread.id)
|
||
for message in msgs['data'][::-1]:
|
||
ans.append(message['content'][0]['text']['value'])
|
||
return ans
|
||
|
||
def rag_assistant(knowledge_name):
|
||
retriever = DashScopeCloudRetriever(knowledge_name)
|
||
pipeline_id = str(retriever.pipeline_id)
|
||
assistant = Assistants.create(
|
||
model='qwen-max',
|
||
name='smart helper',
|
||
description='智能助手，支持知识库查询和插件调用。',
|
||
temperature='0.3',
|
||
instructions="请记住以下材料，他们对回答问题有帮助，请你简洁准确地给出回答，不要给出无关内容。${documents}",
|
||
tools=[
|
||
{
|
||
"type": "code_interpreter"
|
||
},
|
||
{
|
||
"type": "rag",
|
||
"prompt_ra": {
|
||
"pipeline_id": pipeline_id,
|
||
"parameters": {
|
||
"type": "object",
|
||
"properties": {
|
||
"query_word": {
|
||
"type": "str",
|
||
"value": "${documents}"
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}]
|
||
)
|
||
return assistant
|
||
|
||
#TODO:http格式，有bug还没修改
|
||
def create_assistant(knowledge_name):
|
||
"""
|
||
Create an assistant using DashScope API via HTTP request based on the provided knowledge name.
|
||
|
||
Parameters:
|
||
knowledge_name (str): The name of the knowledge base to associate with the assistant.
|
||
|
||
Returns:
|
||
dict: Response from the API containing assistant details.
|
||
|
||
Raises:
|
||
ValueError: If the DASHSCOPE_API_KEY environment variable is not set.
|
||
Exception: If any error occurs during the HTTP request.
|
||
"""
|
||
# Step 1: Initialize the Retriever and get the Pipeline ID
|
||
try:
|
||
retriever = DashScopeCloudRetriever(knowledge_name)
|
||
pipeline_id = str(retriever.pipeline_id)
|
||
except Exception as e:
|
||
print(f"Error retrieving pipeline ID for knowledge '{knowledge_name}': {e}")
|
||
return None
|
||
|
||
# Step 2: Fetch the API Key from Environment Variables
|
||
api_key = os.getenv("DASHSCOPE_API_KEY")
|
||
if not api_key:
|
||
raise ValueError("DASHSCOPE_API_KEY environment variable is not set.")
|
||
|
||
# Step 3: Define the API Endpoint and Headers
|
||
url = 'https://dashscope.aliyuncs.com/api/v1/assistants'
|
||
headers = {
|
||
"Content-Type": "application/json",
|
||
"Authorization": f"Bearer {api_key}"
|
||
}
|
||
|
||
# Step 4: Construct the Instructions
|
||
instructions = (
|
||
"请记住以下材料，他们对回答问题有帮助，请你简洁准确地给出回答，不要给出无关内容。${documents}"
|
||
)
|
||
|
||
# Step 5: Define the Tools
|
||
tools = [
|
||
{
|
||
"type": "code_interpreter"
|
||
},
|
||
{
|
||
"type": "rag",
|
||
"prompt_ra": {
|
||
"pipeline_id": pipeline_id,
|
||
"parameters": {
|
||
"type": "object",
|
||
"properties": {
|
||
"query_word": {
|
||
"type": "str",
|
||
"value": "${documents}"
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
]
|
||
|
||
# Step 6: Construct the Payload
|
||
payload = {
|
||
"model": "qwen-max",
|
||
"name": "智能小助手", # "Smart Helper" in Chinese
|
||
"description": "智能助手，支持知识库查询和插件调用。",
|
||
"temperature": 0.3,
|
||
"instructions": instructions,
|
||
"tools": tools,
|
||
"file_ids": [], # Add file IDs if necessary
|
||
"metadata": {} # Add metadata if necessary
|
||
}
|
||
|
||
# Optional: If you have specific file_ids or metadata, you can modify the payload accordingly
|
||
# For example:
|
||
# payload["file_ids"] = ["file_id_1", "file_id_2"]
|
||
# payload["metadata"] = {"key1": "value1", "key2": "value2"}
|
||
|
||
# Step 7: Make the HTTP POST Request
|
||
try:
|
||
response = requests.post(url, headers=headers, data=json.dumps(payload))
|
||
response.raise_for_status() # Raises HTTPError for bad responses (4xx or 5xx)
|
||
assistant = response.json()
|
||
print("Assistant created successfully:")
|
||
print(json.dumps(assistant, indent=4, ensure_ascii=False))
|
||
return assistant
|
||
except requests.exceptions.HTTPError as http_err:
|
||
print(f"HTTP error occurred: {http_err} - Response: {response.text}")
|
||
except Exception as err:
|
||
print(f"An error occurred: {err}")
|
||
|
||
|
||
def pure_assistant():
|
||
assistant = Assistants.create(
|
||
model='qwen-max',
|
||
name='smart helper',
|
||
description='智能助手，能基于用户的要求精准简洁地回答用户的提问',
|
||
instructions='智能助手，能基于用户的要求精准简洁地回答用户的提问',
|
||
tools=[
|
||
{
|
||
"type": "code_interpreter"
|
||
},
|
||
]
|
||
)
|
||
return assistant
|
||
|
||
def llm_call(question, knowledge_name,file_id, result_queue, ans_index, llm_type):
|
||
if llm_type==1:
|
||
print(f"rag_assistant! question:{question}")
|
||
assistant = rag_assistant(knowledge_name)
|
||
# assistant=create_assistant(knowledge_name)
|
||
elif llm_type==2:
|
||
print(f"qianwen_long! question:{question}")
|
||
qianwen_res = qianwen_long(file_id,question)
|
||
result_queue.put((ans_index,(question,qianwen_res)))
|
||
return
|
||
else :
|
||
assistant = pure_assistant()
|
||
ans = send_message(assistant, message=question)
|
||
result_queue.put((ans_index, (question, ans))) # 在队列中添加索引 (question, ans)
|
||
|
||
def multi_threading(queries, knowledge_name="", file_id="", llm_type=1):
|
||
if not queries:
|
||
return []
|
||
|
||
print("多线程提问：starting multi_threading...")
|
||
result_queue = queue.Queue()
|
||
|
||
# 使用 ThreadPoolExecutor 管理线程
|
||
with concurrent.futures.ThreadPoolExecutor(max_workers=15) as executor:
|
||
# 逐个提交任务，每提交一个任务后休眠1秒
|
||
future_to_query = {}
|
||
for index, query in enumerate(queries):
|
||
future = executor.submit(llm_call, query, knowledge_name, file_id, result_queue, index, llm_type)
|
||
future_to_query[future] = index
|
||
time.sleep(1) # 每提交一个任务后等待1秒
|
||
|
||
# 收集每个线程的结果
|
||
for future in concurrent.futures.as_completed(future_to_query):
|
||
index = future_to_query[future]
|
||
try:
|
||
future.result() # 捕获异常或确认任务完成
|
||
except Exception as exc:
|
||
print(f"Query {index} generated an exception: {exc}")
|
||
# 确保在异常情况下也向 result_queue 添加占位符
|
||
result_queue.put((index, None))
|
||
|
||
# 从队列中获取所有结果并按索引排序
|
||
results = [None] * len(queries)
|
||
while not result_queue.empty():
|
||
index, result = result_queue.get()
|
||
results[index] = result
|
||
# 检查是否所有结果都是 None
|
||
if all(result is None for result in results):
|
||
return []
|
||
# 过滤掉None值
|
||
results = [r for r in results if r is not None]
|
||
# 返回一个保证是列表的结构
|
||
return results
|
||
|
||
if __name__ == "__main__":
|
||
start_time=time.time()
|
||
# # 读取问题列表
|
||
baseinfo_file_path = 'D:\\flask_project\\flask_app\\static\\提示词\\前两章提问总结.txt'
|
||
questions =read_questions_from_file(baseinfo_file_path)
|
||
knowledge_name = "招标解析5word"
|
||
llm_type=1
|
||
results = multi_threading(questions, knowledge_name)
|
||
end_time = time.time()
|
||
if not results:
|
||
print("errror!")
|
||
else:
|
||
print("elapsed time:"+str(end_time-start_time))
|
||
# 打印结果
|
||
for question, response in results:
|
||
print(f"Response: {response}")
|
||
|
||
# file_path = "C:\\Users\\Administrator\\Desktop\\货物标\\zbfiles\\6.2定版视频会议磋商文件(1)\\6.2定版视频会议磋商文件_1-21.pdf"
|
||
# file_id = upload_file(file_path)
|
||
# questions=["该招标文件的项目名称是？项目编号（或招标编号）是？采购人（或招标人）是？采购代理机构（或招标代理机构）是？请按json格式给我提供信息，键名分别是'项目名称','项目编号','采购人','采购代理机构'，若存在未知信息，在对应的键值中填'未知'。","该招标文件的项目概况是？项目基本情况是？请按json格式给我提供信息，键名分别为'项目概况','项目基本情况'，若存在嵌套信息，嵌套内容键名以文件中对应字段命名，而嵌套键值必须与原文保持一致，若存在未知信息，在对应的键值中填'未知'。"]
|
||
# results=multi_threading(questions,"",file_id,2) #1代表使用百炼rag 2代表使用qianwen-long
|
||
# if not results:
|
||
# print("errror!")
|
||
# else:
|
||
# # 打印结果
|
||
# for question, response in results:
|
||
# print(f"Question: {question}")
|
||
# print(f"Response: {response}")
|
||
|
||
# ques=["关于'资格要求',本采购文件第一章第二款要求的内容是怎样的？请按json格式给我提供信息，键名为'资格要求'，而键值需要完全与原文保持一致，不要擅自总结、删减，如果存在未知信息，请在对应键值处填'未知'。"]
|
||
# # ques=["该招标文件的工程名称（项目名称）是？招标编号是？招标人是？招标代理机构是？请按json格式给我提供信息，键名分别是'工程名称','招标编号','招标人','招标代理机构'，若存在未知信息，在对应的键值中填'未知'。","该招标文件的工程概况（或项目概况）是？招标范围是？请按json格式给我提供信息，键名分别为'工程概况','招标范围'，若存在嵌套信息，嵌套内容键名以文件中对应字段命名，若存在未知信息，在对应的键值中填'未知'。"]
|
||
# results = multi_threading(ques, "6.2视频会议docx")
|
||
# if not results:
|
||
# print("errror!")
|
||
# else:
|
||
# # 打印结果
|
||
# for question, response in results:
|
||
# print(f"Question: {question}")
|
||
# print(f"Response: {response}") |