Skip to content

sathwik-21/Content-Creation-using-LLMs

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

<<<<<<< HEAD

    Content Creation with LLM

A project that revolutionizes content creation using Large Language Models (LLMs). This offline system leverages GPT-Neo for text generation, providing flexibility, contextual awareness, and high-quality outputs.

Features:

Text Generation: Generates high-quality, context-aware content. Grammar Feedback: Uses language-tool-python for grammar and style checks. Fully Offline: Runs without internet access, including the GPT-Neo model (~5GB). Customizable Settings: Adjust parameters like max length and temperature for output control. Local Execution: Access the application via your browser on http://127.0.0.1:5000.

System Requirements:

Operating System: Windows 10/11 Disk Space: ~6GB (includes GPT-Neo model and Java runtime) RAM: 8GB or more recommended Dependencies: Python 3.9+ Java Runtime Environment (JRE)

Installation:

Step 1: Extract Files Unzip the provided archive (ContentCreationLLM.zip) into a folder. Ensure the following folders and files are present:

ContentCreationLLM/ ├── ContentCreationLLM.exe ├── models/ │ └── gpt-neo-1.3B/ ├── java/ │ └── bin/ ├── app/ │ ├── templates/ │ └── static/

Step 2: Run the Application Double-click ContentCreationLLM.exe. Open your browser and navigate to: arduino Copy Edit http://127.0.0.1:5000

Step 3: Generate Content Enter a prompt in the provided input box. Adjust Max Length and Temperature as needed. Click Generate to create content.

How It Works:

Text Generation: Uses the GPT-Neo model loaded from the local models/ folder. Grammar Feedback: Integrated language-tool-python checks for grammar and style errors. Java Runtime is used for the LanguageTool backend. Frontend: Flask serves a local web app accessible through your browser.

Configuration:

Modify Model Parameters: To customize model behavior, edit the app/models.py file before creating the .exe: python Copy Edit result = text_generator( prompt, max_length=300, temperature=0.7, top_k=50, top_p=0.9, )

Known Issues:

Startup Issues: Ensure no other service is using port 5000. Java Not Found: Ensure the java/ folder is included and configured correctly.

Credits: Transformers Library: Hugging Face LanguageTool: Open source grammar checker Flask: Python web framework

Content-Creation-using-LLMs

origin/main

About

Developed a web based content generation tool using a pre trained NeoGPT 1.3B language model. Built the application in Python with Flask for the frontend and integrated Hugging Face Transformers, TensorFlow, and NLTK for text processing. Users can input prompts to generate high-quality content, which can be translated into multiple languages.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors