This repository contains some of the common dataset that are used. The datasets available here are -
- Adult (adult.csv)
- Credit (credit.csv)
- Diabetes (diabetes.csv)
- Haberman (haberman.csv)
- Indian Liver Patient (indianLiverPatient.csv)
- Magic (magic.csv)
- Mammographic (mammographic.csv)
- Pulsar (pulsar.csv) - Only data
- Heart Diseases (heart.csv)
- Connectionist Bench (Sonar, Mines vs. Rocks) (sonar.csv)
- SVM guide (svmguide3.csv)
- Liver Disorder (liver_disorder.csv)
- German credit data (german_numer.csv)
- Yearbook dataset (Yearbook/portraits_1905_1954.mat, portraits_1955_1974.mat, portraits_1975_1994.mat, portraits_1995_2013.mat)
- Ecoli (ecoli.csv)
- Forest Cover (forestcov.csv)
- Glass (glass.csv)
- Iris (iris.csv)
- Letter Recognition (letter-recognition.csv)
- Optical Digit Recognition (optdigits.csv)
- Wine Quality (redwine.csv)
- Satellite (satellite.csv)
- Image Segmentation (segment.csv)
- Vehicle Silhouettes (vehicle.csv)
- Pulsar (pulsar.csv) - Only data
The datasets are in the data folder and their description is available in the folder descr
The repo also contains some functions in the file load.py to load these datasets as a numpy matrix. The file example.py gives an example of the usage of these functions. You can run that file to load and see the output of any of these datasets by passing the name of the dataset file as the command line argument -
python example.py datasetname
In order to load a dataset, you can call the corresponding function (load_<datasetname>) available in the file load.py. For example, to load the dataset adult, you need to call the function load_adult(True). Note: you need to pass True as parameter to the function if you want the function to return the dataset and its labels as numpy matrix and vector respectively.
These datasets are taken from the UCI machine learning repository and LIBSVM Data repository