Views
No views yet
pip install autogluon1import pandas as pd
2
3from autogluon.tabular import TabularPredictor
4
5
6if __name__ == '__main__':
7 train_data = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/Inc/train.csv')
8 subsample_size = 5000
9 if subsample_size is not None and subsample_size < len(train_data):
10 train_data = train_data.sample(n=subsample_size, random_state=0)
11 test_data = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/Inc/test.csv')
12
13 tabpfnmix_default = {
14 "model_path_classifier": "autogluon/tabpfn-mix-1.0-classifier",
15 "model_path_regressor": "autogluon/tabpfn-mix-1.0-regressor",
16 "n_ensembles": 1,
17 "max_epochs": 30,
18 }
19
20 hyperparameters = {
21 "TABPFNMIX": [
22 tabpfnmix_default,
23 ],
24 }
25
26 label = "age"
27 problem_type = "regression"
28
29 predictor = TabularPredictor(
30 label=label,
31 problem_type=problem_type,
32 )
33 predictor = predictor.fit(
34 train_data=train_data,
35 hyperparameters=hyperparameters,
36 verbosity=3,
37 )
38
39 predictor.leaderboard(test_data, display=True)@article{erickson2020autogluon,
title={Autogluon-tabular: Robust and accurate automl for structured data},
author={Erickson, Nick and Mueller, Jonas and Shirkov, Alexander and Zhang, Hang and Larroy, Pedro and Li, Mu and Smola, Alexander},
journal={arXiv preprint arXiv:2003.06505},
year={2020}
}
@article{hollmann2022tabpfn,
title={Tabpfn: A transformer that solves small tabular classification problems in a second},
author={Hollmann, Noah and M{\"u}ller, Samuel and Eggensperger, Katharina and Hutter, Frank},
journal={arXiv preprint arXiv:2207.01848},
year={2022}
}
@article{breejen2024context,
title={Why In-Context Learning Transformers are Tabular Data Classifiers},
author={Breejen, Felix den and Bae, Sangmin and Cha, Stephen and Yun, Se-Young},
journal={arXiv preprint arXiv:2405.13396},
year={2024}
}