diff --git a/summerizer_concise.py b/summerizer_concise.py new file mode 100644 index 0000000..30e2d5d --- /dev/null +++ b/summerizer_concise.py @@ -0,0 +1,67 @@ +import argparse + +from langchain.chains.llm import LLMChain +from langchain.chains.summarize import load_summarize_chain +from langchain.prompts import PromptTemplate +from langchain_community.chat_models import ChatOllama +from langchain_community.document_loaders import WebBaseLoader + + +def setup_argparse(): + """Setup argparse to parse command line arguments.""" + parser = argparse.ArgumentParser( + description="Summarize a document from a given URL." + ) + parser.add_argument( + "-u", "--url", required=True, help="URL of the document to summarize" + ) + return parser.parse_args() + + +def load_document(url): + """Load document from the specified URL.""" + loader = WebBaseLoader(url) + return loader.load() + + +def setup_summarization_chain(): + """Setup the summarization chain with a prompt template and ChatOllama.""" + prompt_template = PromptTemplate( + template="""As a professional summarizer, create a concise and comprehensive summary of the provided text, be it an article, post, conversation, or passage, while adhering to these guidelines: + 1. Craft a summary that is detailed, thorough, in-depth, and complex, while maintaining clarity and conciseness. + + 2. Incorporate main ideas and essential information, eliminating extraneous language and focusing on critical aspects. + + 3. Rely strictly on the provided text, without including external information. + + 4. Format the summary in paragraph form for easy understanding. + + 5.Conclude your notes with [End of Notes, Message #X] to indicate completion, where "X" represents the total number of messages that I have sent. In other words, include a message counter where you start with #1 and add 1 to the message counter every time I send a message. + + By following this optimized prompt, you will generate an effective summary that encapsulates the essence of the given text in a clear, concise, and reader-friendly manner. + + Write a short concise summary of the following: + + "{text}" + + CONCISE SUMMARY:""", + input_variables=["text"], + ) + + llm = ChatOllama(model="llama2") + llm_chain = LLMChain(llm=llm, prompt=prompt_template) + return llm_chain + + +def main(): + args = setup_argparse() + docs = load_document(args.url) + + llm_chain = setup_summarization_chain() + result = llm_chain.run(docs) + + print(result) + + +if __name__ == "__main__": + main()