Context-Aware Automated Feature Engineering (CAAFE) is an automated machine learning tool that uses large language models for feature engineering in tabular datasets. It generates Python code for new features along with explanations for their utility, enhancing interpretability.
Project description
Usage
Use this colab notebook for a quickstart.
Use CAFE_minimal.ipynb for a minimal example of how to use CAAFE on your dataset.
Use CAAFE.ipynb to reproduce the experiments from the paper.
Choosing an Iterative Classifier
The iterative classifier gets called in each
Access Tokens
OpenAI
Paper
Hollmann, N., Müller, S., & Hutter, F. (2023). LLMs for Semi-Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineering https://arxiv.org/abs/2305.03403
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.