Enotion Draw User Guide

Introduction 👋

"Emotion Draw" combines state-of-the-art technology, utilizing BERT-based models for natural language processing (NLP) to predict emotions from user input sentences. 🤖💬 In tandem, it leverages Stable Diffusion, a powerful deep learning text-to-image model, to generate expressive images corresponding to the predicted emotions. 🎨✨ This fusion of advanced NLP and image generation techniques enables "Emotion Draw" to provide users with a seamless and immersive experience, bridging the gap between textual and visual expression. 🌟🖼️ Through the integration of these cutting-edge AI capabilities, the platform empowers users to explore and communicate their emotions in entirely new and creative ways. 🚀🎭

Table of Contents 🤓

• Dataset
  • Emotions Dataset for NLP
  • Preprocessing
• 🎨 Emotion Draw with ALBERT: The Mighty Mite! 🤖
  • BERT: The Big, the Bold, and the Brainy - Why We Gave It a Pass! 🫣
  • ALBERT: The Chosen One! 🚀
  • How Do the Others Perform?
• Multiclass Classification Trainer
  • Initialization
  • Functions
• Step-by-Step: Fine-Tuning Bert and Friends
  • Import Packages
  • Choose a Model
  • Specify the Parameters
  • Initialize the Class
  • Load Data
  • Training
  • Evaluate on the Test Set
  • Inference for a Single Example
  • Display Confusion Matrices
  • TensorBoard
• The Artist: Stable Diffusion v2-1 👨🏻‍🎨🎨
  • Introduction to Stable Diffusion Models
  • Overview of Stable Diffusion v2-1 Model
  • Our Mission: Prompt Engineering 💬
• FastAPI Integration
  • Functionality
  • Usage
  • Run the FastAPI Application
  • Example
• Integrating JavaScript and React for the Frontend of Emotion Draw
  • Key Features Implemented
  • Run the JS React Application

Troubleshooting 🎯

If you encounter any issues, reach out to our team.

Contributing 🤝

We welcome contributions! Feel free to submit bug reports, feature requests, or contribute to the codebase on our GitHub repository.

Contact Information 📞

For further assistance or inquiries, contact our support team at

anahit_baghdasaryan2@edu.aua.am vyacheslav_stepanyan@edu.aua.am