Spaces:
Sleeping
Sleeping
title: Feelings To Emoji | |
emoji: π | |
colorFrom: red | |
colorTo: pink | |
sdk: gradio | |
sdk_version: 5.25.0 | |
app_file: app.py | |
pinned: false | |
license: mit | |
short_description: Turn emotions into emoji mashups with embedding models! | |
# Feelings to Emoji | |
This application uses sentence embeddings to match emotional text expressions with appropriate emojis and create unique emoji mashups. | |
## π Features | |
- **Emotion Recognition**: Convert your emotional expressions into matching emojis | |
- **Emoji Mashups**: Generate creative combinations of emotion and event emojis | |
- **Multiple Embedding Models**: Choose from different embedding models to see how they interpret your emotions | |
- **Cached Embeddings**: Pre-computed embeddings for faster response times | |
- **Comprehensive Examples**: Built-in examples based on Plutchik's Wheel of Emotions | |
## π§ Embedding Models | |
Three different sentence embedding models are available: | |
- **MPNet** (110M) - Balanced, great general-purpose model | |
- **GTE** (335M) - Context-rich, good for emotion & nuance | |
- **BGE** (350M) - Tuned for ranking & high-precision similarity | |
## π How It Works | |
1. Enter a sentence describing your feelings or emotions | |
2. The system uses sentence embeddings to match your text with the most similar emotion emoji | |
3. It also finds an event/object emoji that matches your text | |
4. The two emojis are combined into a unique "emoji kitchen" mashup | |
## π Getting Started | |
To run this application locally: | |
1. Clone this repository | |
2. Install requirements: `pip install -r requirements.txt` | |
3. Generate embeddings (optional): `python generate_embeddings.py` | |
4. Run the app: `python app.py` | |
## π References | |
For more details about the implementation and API reference, see [REFERENCE.md](REFERENCE.md). | |