tabular-benchmark / README.md
polinaeterna's picture
polinaeterna HF staff
Update README.md
22a9f49
metadata
annotations_creators: []
license: []
pretty_name: tabular_benchmark
tags: []
task_categories:
  - tabular-classification
  - tabular-regression
configs:
  - config_name: clf_cat_covertype
    data_files: clf_cat/covertype.csv
  - config_name: clf_num_Higgs
    data_files: clf_num/Higgs.csv

Tabular Benchmark

Dataset Description

This dataset is a curation of various datasets from openML and is curated to benchmark performance of various machine learning algorithms.

Dataset Summary

Benchmark made of curation of various tabular data learning tasks, including:

  • Regression from Numerical and Categorical Features
  • Regression from Numerical Features
  • Classification from Numerical and Categorical Features
  • Classification from Numerical Features

Supported Tasks and Leaderboards

  • tabular-regression
  • tabular-classification

Dataset Structure

Data Splits

This dataset consists of four splits (folders) based on tasks and datasets included in tasks.

  • reg_num: Task identifier for regression on numerical features.
  • reg_cat: Task identifier for regression on numerical and categorical features.
  • clf_num: Task identifier for classification on numerical features.
  • clf_cat: Task identifier for classification on categorical features.

Depending on the dataset you want to load, you can load the dataset by passing task_name/dataset_name to data_files argument of load_dataset like below:

from datasets import load_dataset
dataset = load_dataset("inria-soda/tabular-benchmark", data_files="reg_cat/house_sales.csv")

Dataset Creation

Curation Rationale

This dataset is curated to benchmark performance of tree based models against neural networks. The process of picking the datasets for curation is mentioned in the paper as below:

  • Heterogeneous columns. Columns should correspond to features of different nature. This excludes images or signal datasets where each column corresponds to the same signal on different sensors.
  • Not high dimensional. We only keep datasets with a d/n ratio below 1/10.
  • Undocumented datasets We remove datasets where too little information is available. We did keep datasets with hidden column names if it was clear that the features were heterogeneous.
  • I.I.D. data. We remove stream-like datasets or time series.
  • Real-world data. We remove artificial datasets but keep some simulated datasets. The difference is subtle, but we try to keep simulated datasets if learning these datasets are of practical importance (like the Higgs dataset), and not just a toy example to test specific model capabilities.
  • Not too small. We remove datasets with too few features (< 4) and too few samples (< 3 000). For benchmarks on numerical features only, we remove categorical features before checking if enough features and samples are remaining.
  • Not too easy. We remove datasets which are too easy. Specifically, we remove a dataset if a simple model (max of a single tree and a regression, logistic or OLS) reaches a score whose relative difference with the score of both a default Resnet (from Gorishniy et al. [2021]) and a default HistGradientBoosting model (from scikit learn) is below 5%. Other benchmarks use different metrics to remove too easy datasets, like removing datasets perfectly separated by a single decision classifier [Bischl et al., 2021], but this ignores varying Bayes rate across datasets. As tree ensembles are superior to simple trees and logistic regresison [Fernández-Delgado et al., 2014], a close score for the simple and powerful models suggests that we are already close to the best achievable score.
  • Not deterministic. We remove datasets where the target is a deterministic function of the data. This mostly means removing datasets on games like poker and chess. Indeed, we believe that these datasets are very different from most real-world tabular datasets, and should be studied separately

Source Data

Numerical Classification

dataset_name n_samples n_features original_link new_link
electricity 38474.0 7.0 https://www.openml.org/d/151 https://www.openml.org/d/44120
covertype 566602.0 10.0 https://www.openml.org/d/293 https://www.openml.org/d/44121
pol 10082.0 26.0 https://www.openml.org/d/722 https://www.openml.org/d/44122
house_16H 13488.0 16.0 https://www.openml.org/d/821 https://www.openml.org/d/44123
MagicTelescope 13376.0 10.0 https://www.openml.org/d/1120 https://www.openml.org/d/44125
bank-marketing 10578.0 7.0 https://www.openml.org/d/1461 https://www.openml.org/d/44126
Bioresponse 3434.0 419.0 https://www.openml.org/d/4134 https://www.openml.org/d/45019
MiniBooNE 72998.0 50.0 https://www.openml.org/d/41150 https://www.openml.org/d/44128
default-of-credit-card-clients 13272.0 20.0 https://www.openml.org/d/42477 https://www.openml.org/d/45020
Higgs 940160.0 24.0 https://www.openml.org/d/42769 https://www.openml.org/d/44129
eye_movements 7608.0 20.0 https://www.openml.org/d/1044 https://www.openml.org/d/44130
Diabetes130US 71090.0 7.0 https://www.openml.org/d/4541 https://www.openml.org/d/45022
jannis 57580.0 54.0 https://www.openml.org/d/41168 https://www.openml.org/d/45021
heloc 10000.0 22.0 "https://www.kaggle.com/datasets/averkiyoliabev/home-equity-line-of-creditheloc?select=heloc_dataset_v1+%281%29.csv" https://www.openml.org/d/45026
credit 16714.0 10.0 "https://www.kaggle.com/c/GiveMeSomeCredit/data?select=cs-training.csv" https://www.openml.org/d/44089
california 20634.0 8.0 "https://www.dcc.fc.up.pt/ltorgo/Regression/cal_housing.html" https://www.openml.org/d/45028

Categorical Classification

Numerical Regression

dataset_name n_samples n_features original_link new_link
cpu_act 8192.0 21.0 https://www.openml.org/d/197 https://www.openml.org/d/44132
pol 15000.0 26.0 https://www.openml.org/d/201 https://www.openml.org/d/44133
elevators 16599.0 16.0 https://www.openml.org/d/216 https://www.openml.org/d/44134
wine_quality 6497.0 11.0 https://www.openml.org/d/287 https://www.openml.org/d/44136
Ailerons 13750.0 33.0 https://www.openml.org/d/296 https://www.openml.org/d/44137
yprop_4_1 8885.0 42.0 https://www.openml.org/d/416 https://www.openml.org/d/45032
houses 20640.0 8.0 https://www.openml.org/d/537 https://www.openml.org/d/44138
house_16H 22784.0 16.0 https://www.openml.org/d/574 https://www.openml.org/d/44139
delays_zurich_transport 5465575.0 9.0 https://www.openml.org/d/40753 https://www.openml.org/d/45034
diamonds 53940.0 6.0 https://www.openml.org/d/42225 https://www.openml.org/d/44140
Brazilian_houses 10692.0 8.0 https://www.openml.org/d/42688 https://www.openml.org/d/44141
Bike_Sharing_Demand 17379.0 6.0 https://www.openml.org/d/42712 https://www.openml.org/d/44142
nyc-taxi-green-dec-2016 581835.0 9.0 https://www.openml.org/d/42729 https://www.openml.org/d/44143
house_sales 21613.0 15.0 https://www.openml.org/d/42731 https://www.openml.org/d/44144
sulfur 10081.0 6.0 https://www.openml.org/d/23515 https://www.openml.org/d/44145
medical_charges 163065.0 5.0 https://www.openml.org/d/42720 https://www.openml.org/d/44146
MiamiHousing2016 13932.0 14.0 https://www.openml.org/d/43093 https://www.openml.org/d/44147
superconduct 21263.0 79.0 https://www.openml.org/d/43174 https://www.openml.org/d/44148

Categorical Regression

dataset_name n_samples n_features original_link new_link
topo_2_1 8885.0 255.0 https://www.openml.org/d/422 https://www.openml.org/d/45041
analcatdata_supreme 4052.0 7.0 https://www.openml.org/d/504 https://www.openml.org/d/44055
visualizing_soil 8641.0 4.0 https://www.openml.org/d/688 https://www.openml.org/d/44056
delays_zurich_transport 5465575.0 12.0 https://www.openml.org/d/40753 https://www.openml.org/d/45045
diamonds 53940.0 9.0 https://www.openml.org/d/42225 https://www.openml.org/d/44059
Allstate_Claims_Severity 188318.0 124.0 https://www.openml.org/d/42571 https://www.openml.org/d/45046
Mercedes_Benz_Greener_Manufacturing 4209.0 359.0 https://www.openml.org/d/42570 https://www.openml.org/d/44061
Brazilian_houses 10692.0 11.0 https://www.openml.org/d/42688 https://www.openml.org/d/44062
Bike_Sharing_Demand 17379.0 11.0 https://www.openml.org/d/42712 https://www.openml.org/d/44063
Airlines_DepDelay_1M 1000000.0 5.0 https://www.openml.org/d/42721 https://www.openml.org/d/45047
nyc-taxi-green-dec-2016 581835.0 16.0 https://www.openml.org/d/42729 https://www.openml.org/d/44065
abalone 4177.0 8.0 https://www.openml.org/d/42726 https://www.openml.org/d/45042
house_sales 21613.0 17.0 https://www.openml.org/d/42731 https://www.openml.org/d/44066
seattlecrime6 52031.0 4.0 https://www.openml.org/d/42496 https://www.openml.org/d/45043
medical_charges 163065.0 5.0 https://www.openml.org/d/42720 https://www.openml.org/d/45048
particulate-matter-ukair-2017 394299.0 6.0 https://www.openml.org/d/42207 https://www.openml.org/d/44068
SGEMM_GPU_kernel_performance 241600.0 9.0 https://www.openml.org/d/43144 https://www.openml.org/d/44069

Dataset Curators

Léo Grinsztajn, Edouard Oyallon, Gaël Varoquaux.

Licensing Information

[More Information Needed]

Citation Information

Léo Grinsztajn, Edouard Oyallon, Gaël Varoquaux. Why do tree-based models still outperform deep learning on typical tabular data?. NeurIPS 2022 Datasets and Benchmarks Track, Nov 2022, New Orleans, United States. ffhal-03723551v2f