File size: 1,669 Bytes
f5e247b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 |
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain
from langchain_community.chat_models import ChatOpenAI
from langchain_deepseek import ChatDeepSeek
from langchain_core.runnables import RunnableLambda, RunnableMap
from dotenv import load_dotenv
import os
load_dotenv()
def get_summary_prompt():
return PromptTemplate.from_template("Summarize the following:\n\n{text}")
def get_title_prompt():
return PromptTemplate.from_template("Create a 5-word title for this:\n\n{summary}")
def translate_prompt():
return PromptTemplate.from_template("Translate the following text to Chinese:\n\n{text}")
def build_hyper_chain():
llm = ChatDeepSeek(api_key=os.getenv("DEEPSEEK_API_KEY"), model="deepseek-chat", temperature=0.7)
summarize_chain = get_summary_prompt() | llm
title_chain = get_title_prompt() | llm
translate_chain = translate_prompt() | llm
def chain_fn(inputs):
summary = summarize_chain.invoke({"text": inputs["text"]})
title = title_chain.invoke({"summary": summary})
chinese_summary = translate_chain.invoke({"text": summary.content})
return {"summary": summary, "title": title, "chinese_summary": chinese_summary}
return RunnableLambda(chain_fn)
if __name__ == "__main__":
full_text = input("Paste your paragraph:\n\n")
chain = build_hyper_chain()
outputs = chain.invoke({"text": full_text})
# Display outputs
print(f"\n[Summary]: {outputs['summary'].content}")
print(f"\n[Title]: {outputs['title'].content}")
print(f"\n[Chinese Summary]: {outputs['chinese_summary'].content}") |