heart.py

"""Heart"""

from typing import List
from functools import partial

import datasets

import pandas


VERSION = datasets.Version("1.0.0")
_BASE_FEATURE_NAMES = [
    "age",
    "is_male",
    "type_of_chest_pain",
    "resting_blood_pressure",
    "serum_cholesterol",
    "fasting_blood_sugar",
    "rest_electrocardiographic_type",
    "maximum_heart_rate",
    "has_exercise_induced_angina",
    "depression_induced_by_exercise",
    "slope_of_peak_exercise",
    "number_of_major_vessels_colored_by_flourosopy",
    "thal",
    "has_hearth_disease"
]

DESCRIPTION = "Heart dataset from the UCI ML repository."
_HOMEPAGE = "https://archive.ics.uci.edu/ml/datasets/Heart"
_URLS = ("https://huggingface.co/datasets/mstz/heart/raw/heart.csv")
_CITATION = """
@misc{misc_heart_disease_45,
  author       = {Janosi,Andras, Steinbrunn,William, Pfisterer,Matthias, Detrano,Robert & M.D.,M.D.},
  title        = {{Heart Disease}},
  year         = {1988},
  howpublished = {UCI Machine Learning Repository},
  note         = {{DOI}: \\url{10.24432/C52P4X}}
}"""

# Dataset info
urls_per_split = {
    "hungary": {"train": "https://huggingface.co/datasets/mstz/heart/raw/main/processed.hungarian.data"},
}
features_types_per_config = {
    "hungary": {
        "age": datasets.Value("int8"),
        "is_male": datasets.Value("bool"),
        "type_of_chest_pain": datasets.Value("string"),
        "resting_blood_pressure": datasets.Value("float32"),
        "serum_cholesterol": datasets.Value("float32"),
        "fasting_blood_sugar": datasets.Value("float32"),
        "rest_electrocardiographic_type": datasets.Value("string"),
        "maximum_heart_rate": datasets.Value("float32"),
        "has_exercise_induced_angina": datasets.Value("bool"),
        "depression_induced_by_exercise": datasets.Value("float32"),
        "has_hearth_disease": 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}

_ENCODING_DICS = {
    "type_of_chest_pain": {
        1: "typical angina",
        2: "atypical angina",
        3: "non-anginal pain",
        4: "asymptomatic"
    }
}

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


class Heart(datasets.GeneratorBasedBuilder):
    # dataset versions
    DEFAULT_CONFIG = "hungary"
    BUILDER_CONFIGS = [
        HeartConfig(name="hungary",
                    description="Heart for binary classification, hungary dataset.")
    ]


    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[self.config.name]["train"]})
        ]
    
    def _generate_examples(self, filepath: str):
        data = pandas.read_csv(filepath, header=None)
        data.columns = _BASE_FEATURE_NAMES
        data = self.preprocess(data, self.config.name)

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

            yield row_id, data_row

    def preprocess(self, data, config):
        for feature in _ENCODING_DICS:
            encoding_function = partial(self.encode, feature)
            data.loc[:, feature] = data[feature].apply(encoding_function)

        data[["age"]].applymap(int)

        data.drop("slope_of_peak_exercise", axis="columns", inplace=True)
        data.drop("number_of_major_vessels_colored_by_flourosopy", axis="columns", inplace=True)
        data.drop("thal", axis="columns", inplace=True)
        data = data[data.serum_cholesterol != "?"]

        data = data.infer_objects()

        data = data[data.resting_blood_pressure != "?"]
        data = data[data.fasting_blood_sugar != "?"]
        data = data[data.rest_electrocardiographic_type != "?"]
        data = data[data.maximum_heart_rate != "?"]
        data = data[data.has_exercise_induced_angina != "?"]

        data = data.astype({"is_male": bool, "has_exercise_induced_angina": bool,
                            "serum_cholesterol": float, "maximum_heart_rate": float,
                            "resting_blood_pressure": float, "fasting_blood_sugar": float})       

        return data

    def encode(self, feature, value):
        if feature in _ENCODING_DICS:
            return _ENCODING_DICS[feature][value]
        raise ValueError(f"Unknown feature: {feature}")