pima.py

from typing import List

import datasets

import pandas


VERSION = datasets.Version("1.0.0")


DESCRIPTION = "Pima dataset from the OpenML repository."
_HOMEPAGE = "https://www.openml.org/search?type=data&status=active&id=37"
_URLS = ("https://www.openml.org/search?type=data&status=active&id=37")
_CITATION = """"""

# Dataset info
urls_per_split = {
    "train": "https://huggingface.co/datasets/mstz/pima/raw/main/pima.csv"
}
features_types_per_config = {
    "pima": {
	    "number_of_pregnancies": datasets.Value("int8"),
		"plasma_glucose_concentration": datasets.Value("float64"),
		"diastolic_blood_pressure": datasets.Value("float64"),
		"triceps_thickness": datasets.Value("float64"),
		"serum_insulin": datasets.Value("float64"),
		"bmi": datasets.Value("float64"),
		"diabetes_pedigree": datasets.Value("float64"),
		"age": datasets.Value("float64"),
		"has_diabetes": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
    },
    
}
features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}


class PimaConfig(datasets.BuilderConfig):
    def __init__(self, **kwargs):
        super(PimaConfig, self).__init__(version=VERSION, **kwargs)
        self.features = features_per_config[kwargs["name"]]


class Pima(datasets.GeneratorBasedBuilder):
    # dataset versions
    DEFAULT_CONFIG = "pima"
    BUILDER_CONFIGS = [
        PimaConfig(name="pima",
                    description="Pima for binary classification."),
        ]


    def _info(self):       
        info = datasets.DatasetInfo(description=DESCRIPTION, citation=_CITATION, homepage=_HOMEPAGE,
                                    features=features_per_config[self.config.name])

        return info
    
    def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
        downloads = dl_manager.download_and_extract(urls_per_split)

        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloads["train"]})
        ]
    
    def _generate_examples(self, filepath: str):
        data = pandas.read_csv(filepath)
        data = data.rename(columns={"class": "has_diabetes"})

        for row_id, row in data.iterrows():
            data_row = dict(row)

            yield row_id, data_row