Pypercraft

A Novel Application that Harnesses LLMs to Generate Articles and Papers

Abstract

Introducing a novel Python application that harnesses the expanded capabilities of ChatGPT, enabling users to effortlessly compose comprehensive articles and papers on topics of their choice. Surpassing token constraints, this application empowers users to craft full compositions comprising a Title, Introduction, Body, and Conclusion. Accessible through a user-friendly Streamlit interface or a FastAPI-powered API, the tool seamlessly merges user input with ChatGPT's contextual understanding, resulting in coherent written pieces. Offering an additional advantage, users can export their creations as Word documents (.docx), enhancing the versatility and utility of the generated content. Whether fashioning educational articles or structured research papers, this application revolutionizes content creation, amalgamating user ingenuity with ChatGPT's linguistic finesse and streamlining written communication with convenience and sophistication.

Live Demo

A live demo of the Streamlit application can be seen here

Quick Start Guide

Clone The Repo:

git clone git@github.com:alkhalifas/pypercraft.git

pip install -r requirements.txt

streamlit run ui.py

Install With Pip:

Install Library:

pip install pypercraft

Import and Apply:

from pypercraft import pypercraft

craft = pypercraft.Pypercraft(

        # Describe the paper you want to generate
        query= "A Scientific Paper about Deep Learning",

        # Select the topic of your paper
        topic= "Data Science",

        # Select number of pages
        num_pages= 3,

        # Select tone of the writer
        tone= "professional",

        # Enter API key
        api_key= os.getenv("OPENAI_API_KEY"))

# Construct the paper
paper = craft.construct()

# Export final paper
craft.export_docx('mypaper.docx')