HeraMod

LLM designed to help, motivate (or demotivate) people (take responses with a grain of salt).

View Demo · Report Bug · Request Feature

Contents

1. About The Project
   • Built With
2. Getting Started
   • Prerequisites
   • Installation
3. Usage
4. Contributing
5. License
6. Contact
7. Acknowledgments

About The Project

HeraMod is an LLM that was trained on couple of self-motivational, self-improvement related texts. It is supposed to generate kind, motivating words, that might help everyone.

HeraMod is short of Hera Model, Hera is goddess of:

• Childbirth - which is considered something awesome, giving new life is a hard, yet rewarding task - this LLM generates amazing words based on hard prompts
• Family - we feel safe around family, can do anything thanks to them. HeraMod is supposed to give you the same feeling - you are limited only by your own mind!
• Marriage - it is something that influences our entire lifes. Same goes for this model - a small answer might change someone's experience with life entirely.

It isn't yet written in the best way, it's made for scratch - no pretrained weights. Thus responses might not be sometimes adequate to the topic. It will be trained further and improved.

(Top)

Built With

• 
• 

(Top)

Getting Started

Here is an example on how to set up your project locally

Prerequisites

Install all requirements from requirements.txt

Running

Run your program in any Python IDE or editor of your liking use:

• App.py - for Gradio interface
• gptGenerate - to generate a prompt from terminal
• gptTrain - to train Model on text files from resources/data directory

(Top)

Usage

Here are example prompts (LLM will finish them):

(Top)