OpenML
Filter results by:
A family of datasets synthetically generated from a simulation of how bank-customers choose their banks. Tasks are based on predicting the fraction of bank customers who leave the bank because of full…
0 runs0 likes1 downloads1 reach5 impact
8192 instances - 33 features - 0 classes - 0 missing values
No data.
0 runs0 likes1 downloads1 reach1 impact
1000000 instances - 33 features - 0 classes - 0 missing values
A dataset of steel plates' faults, classified into 7 different types. The goal was to train machine learning for automatic pattern recognition. The dataset consists of 27 features describing each…
277313 runs1 likes35 downloads36 reach16 impact
1941 instances - 34 features - 2 classes - 0 missing values
This is one of a family of datasets synthetically generated from a realistic simulation of the dynamics of a Unimation Puma 560 robot arm. There are eight datastets in this family . In this repository…
0 runs0 likes5 downloads5 reach5 impact
8192 instances - 33 features - 0 classes - 0 missing values
No data.
0 runs0 likes1 downloads1 reach1 impact
1000000 instances - 33 features - 0 classes - 0 missing values
Abstract: This data set contains a total 5820 evaluation scores provided by students from Gazi University in Ankara (Turkey). There is a total of 28 course specific questions and additional 5…
0 runs0 likes2 downloads2 reach5 impact
5820 instances - 33 features - classes - 0 missing values
1. Title: Wisconsin Prognostic Breast Cancer (WPBC) 2. Source Information a) Creators: Dr. William H. Wolberg, General Surgery Dept., University of Wisconsin, Clinical Sciences Center, Madison, WI…
5 runs0 likes4 downloads4 reach1 impact
194 instances - 33 features - 0 classes - 0 missing values
This is one of 41 drug design datasets. The datasets with 1143 features are formed using Adriana.Code software (www.molecular-networks.com/software/adrianacode). The molecules and outputs are taken…
0 runs0 likes2 downloads2 reach5 impact
195 instances - 33 features - 0 classes - 0 missing values
Regroups information for about 7800 different US colleges. Including geographical information, stats about the population attending and post graduation career earnings.
0 runs0 likes0 downloads0 reach0 impact
7063 instances - 50 features - 0 classes - 125494 missing values
Binarized version of the original data set (see version 1). The multi-class target feature is converted to a two-class nominal target feature by re-labeling the majority class as positive ('P') and…
744 runs0 likes8 downloads8 reach8 impact
7019 instances - 61 features - 2 classes - 43814 missing values
Datasets from ACM KDD Cup (http://www.sigkdd.org/kddcup/index.php) Data set for KDD Cup 1999 Modified by TunedIT (converted to ARFF format)…
4 runs0 likes19 downloads19 reach6 impact
4898431 instances - 42 features - 23 classes - 0 missing values
This is one of 41 drug design datasets. The datasets with 1143 features are formed using Adriana.Code software (www.molecular-networks.com/software/adrianacode). The molecules and outputs are taken…
0 runs0 likes1 downloads1 reach5 impact
16 instances - 34 features - 0 classes - 0 missing values
This is a 10% stratified subsample of the data from the 1999 ACM KDD Cup (http://www.sigkdd.org/kddcup/index.php). Modified by TunedIT (converted to ARFF format)…
25 runs1 likes35 downloads36 reach7 impact
494020 instances - 42 features - 23 classes - 0 missing values
This radar data was collected by a system in Goose Bay, Labrador. This system consists of a phased array of 16 high-frequency antennas with a total transmitted power on the order of 6.4 kilowatts. See…
2484 runs3 likes27 downloads30 reach2 impact
351 instances - 35 features - 2 classes - 0 missing values
2126 fetal cardiotocograms (CTGs) were automatically processed and the respective diagnostic features measured. The CTGs were also classified by three expert obstetricians and a consensus…
24175 runs4 likes27 downloads31 reach48 impact
2126 instances - 36 features - 10 classes - 0 missing values
A 3-class version of Cardiotocography dataset.
134 runs0 likes13 downloads13 reach6 impact
2126 instances - 36 features - 3 classes - 0 missing values
pie chart 2
101 runs0 likes5 downloads5 reach5 impact
745 instances - 37 features - 2 classes - 0 missing values
No data.
292 runs0 likes4 downloads4 reach2 impact
1000000 instances - 37 features - 6 classes - 0 missing values
The satellite dataset comprises of features extracted from satellite observations. In particular, each image was taken under four different light wavelength, two in visible light (green and red) and…
2074 runs3 likes67 downloads70 reach21 impact
5100 instances - 37 features - 2 classes - 0 missing values
Multivariate regression data set from: https://link.springer.com/article/10.1007%2Fs10994-016-5546-z : The Andromeda dataset (Hatzikos et al. 2008) concerns the prediction of future values for six…
0 runs0 likes0 downloads0 reach1 impact
49 instances - 36 features - classes - 0 missing values
Multivariate regression data set from: https://link.springer.com/article/10.1007%2Fs10994-016-5546-z : The Andromeda dataset (Hatzikos et al. 2008) concerns the prediction of future values for six…
0 runs0 likes0 downloads0 reach1 impact
49 instances - 36 features - classes - 0 missing values
The database consists of the multi-spectral values of pixels in 3x3 neighbourhoods in a satellite image, and the classification associated with the central pixel in each neighbourhood. The aim is to…
27180 runs2 likes23 downloads25 reach2 impact
6430 instances - 37 features - 6 classes - 0 missing values
One of the NASA Metrics Data Program defect data sets. Data from flight software for earth orbiting satellite. Data comes from McCabe and Halstead features extractors of source code. These features…
875 runs0 likes12 downloads12 reach8 impact
5589 instances - 37 features - 2 classes - 0 missing values
pie chart 1
102 runs0 likes5 downloads5 reach5 impact
705 instances - 38 features - 2 classes - 0 missing values
pie chart 3
103 runs0 likes6 downloads6 reach5 impact
1077 instances - 38 features - 2 classes - 0 missing values
Mega watt
183 runs0 likes8 downloads8 reach7 impact
253 instances - 38 features - 2 classes - 0 missing values
Pizza cutter 3
188 runs0 likes6 downloads6 reach6 impact
1043 instances - 38 features - 2 classes - 0 missing values
Costa madre 1
90 runs0 likes6 downloads6 reach7 impact
296 instances - 38 features - 2 classes - 0 missing values
cast metal 1
111 runs0 likes8 downloads8 reach5 impact
327 instances - 38 features - 2 classes - 0 missing values
* Dataset Title: AutoUniv Dataset data problem: autoUniv-au6-250-drift-au6-cd1-500 * Abstract: AutoUniv is an advanced data generator for classifications tasks. The aim is to reflect the nuances and…
11011 runs0 likes9 downloads9 reach39 impact
750 instances - 41 features - 8 classes - 0 missing values
* Dataset Title: AutoUniv Dataset data problem: autoUniv-au6-cd1-400 * Abstract: AutoUniv is an advanced data generator for classifications tasks. The aim is to reflect the nuances and heterogeneity…
144 runs0 likes3 downloads3 reach5 impact
400 instances - 41 features - 8 classes - 0 missing values
* Dataset Title: AutoUniv Dataset data problem: autoUniv-au6-1000 * Abstract: AutoUniv is an advanced data generator for classifications tasks. The aim is to reflect the nuances and heterogeneity of…
11010 runs0 likes16 downloads16 reach39 impact
1000 instances - 41 features - 8 classes - 0 missing values
No data.
0 runs0 likes2 downloads2 reach1 impact
1000000 instances - 37 features - 0 classes - 0 missing values
Source: Ashwin Srinivasan Department of Statistics and Data Modeling University of Strathclyde Glasgow Scotland UK ross '@' uk.ac.turing The original Landsat data for this database was generated from…
1 runs1 likes6 downloads7 reach7 impact
6435 instances - 37 features - 0 classes - 0 missing values
Pizza cutter
197 runs0 likes8 downloads8 reach6 impact
661 instances - 38 features - 2 classes - 0 missing values
Mean While 1
0 runs0 likes3 downloads3 reach3 impact
253 instances - 38 features - 2 classes - 0 missing values
One of the NASA Metrics Data Program defect data sets. Data from flight software for earth orbiting satellite. Data comes from McCabe and Halstead features extractors of source code. These features…
144328 runs1 likes16 downloads17 reach18 impact
1563 instances - 38 features - 2 classes - 0 missing values
One of the NASA Metrics Data Program defect data sets. Data from flight software for earth orbiting satellite. Data comes from McCabe and Halstead features extractors of source code. These features…
113214 runs0 likes15 downloads15 reach18 impact
1458 instances - 38 features - 2 classes - 0 missing values
%-*- text -*- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% This is a PROMISE data set made publicly available in order to encourage repeatable, verifiable, refutable,…
765 runs0 likes9 downloads9 reach7 impact
403 instances - 38 features - 2 classes - 0 missing values
Ask a home buyer to describe their dream house, and they probably won't begin with the height of the basement ceiling or the proximity to an east-west railroad. But this playground competition's…
0 runs0 likes0 downloads0 reach0 impact
1460 instances - 81 features - 0 classes - 6965 missing values
One of the NASA Metrics Data Program defect data sets. The specific type of software is unknown. Data comes from McCabe and Halstead features extractors of source code. These features were defined in…
815 runs0 likes14 downloads14 reach9 impact
9466 instances - 39 features - 2 classes - 0 missing values
Kung chi
1 runs0 likes4 downloads4 reach4 impact
123 instances - 40 features - 2 classes - 0 missing values
knugget chase 3
0 runs0 likes2 downloads2 reach4 impact
194 instances - 40 features - 2 classes - 0 missing values
Mind Cave 2
0 runs0 likes3 downloads3 reach3 impact
125 instances - 40 features - 2 classes - 0 missing values
One of the NASA Metrics Data Program defect data sets. The specific type of software is unknown. Data comes from McCabe and Halstead features extractors of source code. These features were defined in…
772 runs0 likes10 downloads10 reach7 impact
161 instances - 40 features - 2 classes - 0 missing values
One of the NASA Metrics Data Program defect data sets. The specific type of software is unknown. Data comes from McCabe and Halstead features extractors of source code. These features were defined in…
777 runs0 likes9 downloads9 reach7 impact
458 instances - 40 features - 2 classes - 0 missing values
This is one of 41 drug design datasets. The datasets with 1143 features are formed using Adriana.Code software (www.molecular-networks.com/software/adrianacode). The molecules and outputs are taken…
0 runs0 likes0 downloads0 reach5 impact
22 instances - 40 features - 0 classes - 0 missing values
No data.
307 runs0 likes3 downloads3 reach2 impact
1000000 instances - 41 features - 3 classes - 0 missing values
####1. Summary This database was generated by the Laboratory of Image Processing and Pattern Recognition (INPG-LTIRF) in the development of the Esprit project ELENA No. 6891 and the Esprit working…
18284 runs0 likes11 downloads11 reach10 impact
5500 instances - 41 features - 11 classes - 0 missing values
__Major changes w.r.t. version 1: changed binary features to data type factor.__ Dataset from the Agnostic Learning vs. Prior Knowledge Challenge (http://www.agnostic.inf.ethz.ch), which consisted of…
0 runs0 likes0 downloads0 reach2 impact
14395 instances - 217 features - classes - 0 missing values
Generator generating 3 classes of waves. Each class is generated from a combination of 2 of 3 "base" waves. For details, see Breiman,L., Friedman,J.H., Olshen,R.A., and Stone,C.J. (1984).…
19670 runs1 likes53 downloads54 reach2 impact
5000 instances - 41 features - 3 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
602 runs0 likes12 downloads12 reach7 impact
13750 instances - 41 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). The multi-class target feature is converted to a two-class nominal target feature by re-labeling the majority class as positive ('P') and…
778 runs0 likes9 downloads9 reach7 impact
5000 instances - 41 features - 2 classes - 0 missing values
QSAR biodegradation Data Set * Abstract: Data set containing values for 41 attributes (molecular descriptors) used to classify 1055 chemicals into 2 classes (ready and not ready biodegradable). *…
265103 runs1 likes17 downloads18 reach18 impact
1055 instances - 42 features - 2 classes - 0 missing values
No data.
0 runs0 likes0 downloads0 reach1 impact
1000000 instances - 41 features - 0 classes - 0 missing values
One of two multivariate regression data sets from paper industry, from an experiment at the paper plant Saugbruksforeningen, Norway. They have been described and analysed in: Aldrin, M. (1996),…
0 runs0 likes0 downloads0 reach5 impact
30 instances - 41 features - 0 classes - 0 missing values
This data set addresses a control problem, namely flying a F16 aircraft. The attributes describe the status of the aeroplane, while the goal is to predict the control action on the ailerons of the…
0 runs0 likes6 downloads6 reach5 impact
13750 instances - 41 features - 0 classes - 0 missing values
)), [PMLB](https://github.com/EpistasisLab/penn-ml-benchmarks/tree/master/datasets/classification/tokyo1) This is Performance co-pilot (PCP) data for the Tokyo server at Silicon Graphics International…
35 runs0 likes1 downloads1 reach12 impact
959 instances - 45 features - 2 classes - 0 missing values
This is a corrected version of the previous data file in version 1, which contained a dataset (349 instances) incorrectly merged from the original training and test sets available on UCI (there are…
0 runs0 likes2 downloads2 reach4 impact
267 instances - 45 features - 2 classes - 0 missing values
No data.
43 runs0 likes2 downloads2 reach1 impact
1000000 instances - 45 features - 2 classes - 0 missing values
No data.
47 runs0 likes1 downloads1 reach1 impact
1000000 instances - 45 features - 2 classes - 0 missing values
SPECTF heart data This is a merged version of the separate train and test set which are usually distributed. On OpenML this train-test split can be found as one of the possible tasks. NOTE: See the…
1103 runs0 likes12 downloads12 reach7 impact
349 instances - 45 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
729 runs0 likes9 downloads9 reach6 impact
45 instances - 47 features - 2 classes - 0 missing values
These data are estimated correlations between daily 3 p.m. wind measurements during September and October 1997 for a network of 45 stations in the Sydney region. The first column below gives a list of…
0 runs0 likes0 downloads0 reach3 impact
45 instances - 47 features - classes - 0 missing values
No data.
52 runs0 likes3 downloads3 reach2 impact
1000000 instances - 48 features - 10 classes - 0 missing values
One of a set of 6 datasets describing features of handwritten numerals (0 - 9) extracted from a collection of Dutch utility maps. Corresponding patterns in different datasets correspond to the same…
32605 runs0 likes21 downloads21 reach2 impact
2000 instances - 48 features - 10 classes - 0 missing values
Binarized version of the original data set (see version 1). The multi-class target feature is converted to a two-class nominal target feature by re-labeling the majority class as positive ('P') and…
792 runs0 likes8 downloads8 reach7 impact
2000 instances - 48 features - 2 classes - 0 missing values
The goal of this challenge is to expose the research community to real world datasets of interest to 4Paradigm. All datasets are formatted in a uniform way, though the type of data might differ. The…
0 runs0 likes0 downloads0 reach6 impact
4147 instances - 49 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
624 runs0 likes10 downloads10 reach7 impact
15000 instances - 49 features - 2 classes - 0 missing values
This is a commercial application described in Weiss & Indurkhya (1995). The data describes a telecommunication problem. No further information is available. Characteristics: (10000+5000) cases, 49…
2 runs0 likes3 downloads3 reach1 impact
15000 instances - 49 features - 0 classes - 0 missing values
Oil dataset Past Usage: 1. Kubat, M., Holte, R.,
200 runs3 likes17 downloads20 reach15 impact
937 instances - 50 features - 2 classes - 0 missing values
This dataset is taken from the MiniBooNE experiment and is used to distinguish electron neutrinos (signal) from muon neutrinos (background). This dataset is ordered. It first contains all signal…
6 runs0 likes3 downloads3 reach4 impact
130064 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
766 runs0 likes7 downloads7 reach6 impact
100 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
748 runs0 likes6 downloads6 reach7 impact
250 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
781 runs0 likes8 downloads8 reach7 impact
500 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
755 runs0 likes6 downloads6 reach7 impact
250 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
620 runs0 likes10 downloads10 reach7 impact
1000 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
621 runs0 likes8 downloads8 reach7 impact
1000 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
784 runs0 likes7 downloads7 reach7 impact
500 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
614 runs0 likes9 downloads9 reach7 impact
1000 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
646 runs0 likes9 downloads9 reach7 impact
1000 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
764 runs0 likes6 downloads6 reach6 impact
100 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
636 runs0 likes8 downloads8 reach7 impact
1000 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
810 runs0 likes6 downloads6 reach6 impact
100 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
773 runs0 likes6 downloads6 reach6 impact
100 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
772 runs0 likes7 downloads7 reach7 impact
500 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
817 runs0 likes7 downloads7 reach7 impact
250 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
775 runs0 likes6 downloads6 reach7 impact
250 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
788 runs0 likes7 downloads7 reach6 impact
100 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
786 runs0 likes6 downloads6 reach7 impact
250 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
801 runs0 likes8 downloads8 reach7 impact
500 instances - 51 features - 2 classes - 0 missing values
Binarized version of the original data set (see version 1). It converts the numeric target feature to a two-class nominal target feature by computing the mean and classifying all instances with a…
807 runs0 likes7 downloads7 reach7 impact
500 instances - 51 features - 2 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes1 downloads1 reach5 impact
1000 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes0 downloads0 reach5 impact
100 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes1 downloads1 reach5 impact
1000 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes0 downloads0 reach5 impact
100 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes0 downloads0 reach5 impact
500 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes0 downloads0 reach5 impact
250 instances - 51 features - 0 classes - 0 missing values
The Friedman datasets are 80 artificially generated datasets originating from: J.H. Friedman (1999). Stochastic Gradient Boosting The dataset names are coded as…
0 runs0 likes0 downloads0 reach5 impact
500 instances - 51 features - 0 classes - 0 missing values
Source: James P Bridge, Sean B Holden and Lawrence C Paulson University of Cambridge Computer Laboratory William Gates Building 15 JJ Thomson Avenue Cambridge CB3 0FD UK +44 (0)1223 763500…
24391 runs1 likes20 downloads21 reach34 impact
6118 instances - 52 features - 6 classes - 0 missing values