A tool to generate an emoji based on an emotion, sentence, phrase or even a single word. Implements simple RAG on the llm-emoji-dataset by using MongoDB Atlas Vector search for semantic similarity.
The emoji in the header was generated by
RAGojitoo!
Clone the GitHub repository and navigate to the root folder.
Install the necessary dependencies.
python -m pip install -r requirements.txtNow run app.py using this command.
uvicorn app:app --reload --port 8000Open a browser and go to http://localhost:8000/docs to test the endpoint.
Just install torch for CPU to get started.
python -m pip install torchcuda is recommended for a faster inference if you have a GPU available.
Install the NVIDIA CUDA Toolkit (version greater than or equal to 11.8) and torch >= 11.8.
Note: Make sure the toolkit and
torchhave same versions (i.e., 11.8)
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118Now run the following command in a Python shell to confirm installation.
>> import torch
>> print(torch.cuda.is_available())
TrueI would like to acknowledge the owner of the llm-emoji-dataset.
Feel free to open up a pull request or create an issue.