DeepTLF significantly outperforms traditional Deep Neural Networks (DNNs) in handling tabular data. Using our novel TreeDrivenEncoder, we transform complex, heterogeneous data into a format highly compatible with DNNs. This enables a 19.6% average performance increase compared to conventional DNNs.
You can install DeepTLF directly from PyPI:
pip install deeptlf
Seamlessly integrate DeepTLF into your workflow through its scikit-learn-compatible API:
from deeptlf import DeepTFL
# Initialize and train model
dtlf_model = DeepTFL(n_est=23, max_depth=3, drop=0.23, n_layers=4, task='class')
dtlf_model.fit(X_train, y_train)
# Make predictions
dtlf_y_hat = dtlf_model.predict(X_test)
- Transforms heterogeneous data into DNN-friendly format
- Supports multimodal learning
- Adheres to the scikit-learn API for effortless integration
- Features advanced options like custom layers, dropout rates, and more
To cite DeepTLF in your work:
@article{borisov2022deeptlf,
title={DeepTLF: robust deep neural networks for heterogeneous tabular data},
author={Borisov, Vadim and Broelemann, Klaus and Kasneci, Enkelejda and Kasneci, Gjergji},
journal={International Journal of Data Science and Analytics},
pages={1--16},
year={2022},
publisher={Springer}
}