Embeddings Visualizer is a Python package that provides tools for visualizing text embeddings generated from the OpenAI API. The package uses Streamlit for creating interactive visualization dashboards, and can be executed within a local Python environment or deployed to a web server.
Initialize the package using the init command:
ev init
This will guide you through the process of setting up your configuration. Please ensure you have your OpenAI API key available.
Once initialized, you can start the Streamlit application using the start-app command:
ev start-app
This will start the Streamlit application where you can upload your dataset and interactively visualize the embeddings.
To open the notebook, use the open-notebook command:
ev open-notebook
This will open the embeddings.ipynb Jupyter notebook in your browser, where you can interactively experiment with generating and visualizing embeddings.
Python 3.9 or later is required to use this package. It also depends on several Python libraries, including Streamlit, Typer, numpy, pandas, OpenAI, python-dotenv, scikit-learn, plotly, matplotlib and langchain. These dependencies are automatically installed when you install the Embeddings Visualizer package.
For more information, please refer to the pyproject.toml file in the root directory of the project.