LQAgentPlatform/utils_test/RE_Rrank_Test/debug_rerank.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

"""
调试rerank请求差异
"""
import requests
import json

url = "http://192.168.91.253:9004/v1/rerank"

prefix = '<|im_start|>system\nJudge whether the Document meets the requirements based on the Query and the Instruct provided. Note that the answer can only be "yes" or "no".<|im_end|>\n<|im_start|>user\n'
suffix = "<|im_end|>\n<|im_start|>assistant\n"

query_template = "{prefix}<Instruct>: {instruction}\n<Query>: {query}\n"
document_template = "<Document>: {doc}{suffix}"

instruction = "Given a web search query, retrieve relevant passages that answer the query"

query = "什么是大语言模型？"
candidates = [
    "Qwen 是阿里巴巴推出的大模型系列。",
    "今天天气很好，适合出去玩。",
    "大语言模型是一种基于Transformer的模型",
    "vegetable"
]

# 格式化
formatted_query = query_template.format(prefix=prefix, instruction=instruction, query=query)
formatted_documents = [document_template.format(doc=doc, suffix=suffix) for doc in candidates]

print("=" * 80)
print("请求数据:")
print("=" * 80)
request_data = {
    "query": formatted_query,
    "documents": formatted_documents
}
print(json.dumps(request_data, ensure_ascii=False, indent=2))

print("\n" + "=" * 80)
print("发送请求...")
print("=" * 80)
response = requests.post(url, json=request_data).json()

print("\n" + "=" * 80)
print("响应结果:")
print("=" * 80)
print(json.dumps(response, ensure_ascii=False, indent=2))

print("\n" + "=" * 80)
print("关键信息:")
print("=" * 80)
if "results" in response:
    for idx, result in enumerate(response["results"]):
        print(f"排名{idx+1}: index={result['index']}, score={result['relevance_score']:.6f}")
        print(f"  文档预览: {result['document']['text'][:50]}...")