admin管理员组

文章数量:1355535

from langchain.schema.runnable import RunnableParallel, RunnableLambda
import json
import re

# Define retrieval function
def retrieve_context(inputs):
    query = f"{inputs['subject']} - {', '.join(inputs['topics'])}"  # Combine subject & topic for better retrieval
    retrieved_docs = retriever.invoke(query)
    print(retrieved_docs)
    if not retrieved_docs:
        return "No relevant documents found."

    return "\n\n".join([doc.page_content for doc in retrieved_docs])

rag_chain = (
    RunnableParallel({
        "subject": lambda x: x["subject"],
        "topics": lambda x: x["topics"],
        "context": RunnableLambda(retrieve_context),
        #"response_json":lambda x: json.dumps(RESPONSE_JSON)
    })
    | mcq_prompt
    | llm
)

# Invoke the RAG Chain
response = rag_chain.invoke({"subject": "Chemistry", "topics": ["structure of benzene","From carbonyl compounds"," AromAtic HydrocArbon"]})

# Extract the JSON part using regex
json_match = re.search(r"\[\s*{.*}\s*\]", response.content, re.DOTALL)  # Find JSON array in response
if json_match:
    json_str = json_match.group(0)  # Extract only JSON content
else:
    print("Error: JSON not found in response")
    json_str = "[]"

# Parse JSON
try:
    mcq_data = json.loads(json_str)
except json.JSONDecodeError as e:
    print("Error parsing JSON:", e)
    mcq_data = []

# Store in a structured dictionary
mcq_dict = {f"Q{i+1}": item for i, item in enumerate(mcq_data)}
strong text
print(json.dumps(mcq_dict, indent=2))

The above code is LLM model generate the MCQ questions using Rag method using langchain framework, I am struggle in the part of displaying the generated question to the frontend. How to connect this code with the flask framework for displaying the questions

本文标签：

版权声明：本文标题：How to access the function in jupyter notebook in python language through the flask framework for displaying the generated mcq q 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.betaflare.com/web/1744050863a2582395.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。