covertype.py

from typing import List

import datasets

import pandas


VERSION = datasets.Version("1.0.0")


DESCRIPTION = "Covertype dataset from the UCI ML repository."
_HOMEPAGE = "https://archive-beta.ics.uci.edu/dataset/31/covertype"
_URLS = ("https://archive-beta.ics.uci.edu/dataset/31/covertype")
_CITATION = """"""

# Dataset info
urls_per_split = {
    "train": "https://huggingface.co/datasets/mstz/covertype/raw/main/covtype.data"
}
features_types_per_config = {
    "covertype": {
        "elevation": datasets.Value("float32"),
		"aspect": datasets.Value("float32"),
		"slope": datasets.Value("float32"),
		"horizontal_distance_to_hydrology": datasets.Value("float32"),
		"vertical_distance_to_hydrology": datasets.Value("float32"),
		"horizontal_distance_to_roadways": datasets.Value("float32"),
		"hillshade_9am": datasets.Value("float32"),
		"hillshade_noon": datasets.Value("float32"),
		"hillshade_3pm": datasets.Value("float32"),
		"horizontal_distance_to_fire_points": datasets.Value("float32"),
		"is_a_wilderness_area": datasets.Value("bool"),
		"soil_type_id_0": datasets.Value("bool"),
		"soil_type_id_1": datasets.Value("bool"),
		"soil_type_id_2": datasets.Value("bool"),
		"soil_type_id_3": datasets.Value("bool"),
		"soil_type_id_4": datasets.Value("bool"),
		"soil_type_id_5": datasets.Value("bool"),
		"soil_type_id_6": datasets.Value("bool"),
		"soil_type_id_7": datasets.Value("bool"),
		"soil_type_id_8": datasets.Value("bool"),
		"soil_type_id_9": datasets.Value("bool"),
		"soil_type_id_10": datasets.Value("bool"),
		"soil_type_id_11": datasets.Value("bool"),
		"soil_type_id_12": datasets.Value("bool"),
		"soil_type_id_13": datasets.Value("bool"),
		"soil_type_id_14": datasets.Value("bool"),
		"soil_type_id_15": datasets.Value("bool"),
		"soil_type_id_16": datasets.Value("bool"),
		"soil_type_id_17": datasets.Value("bool"),
		"soil_type_id_18": datasets.Value("bool"),
		"soil_type_id_19": datasets.Value("bool"),
		"soil_type_id_20": datasets.Value("bool"),
		"soil_type_id_21": datasets.Value("bool"),
		"soil_type_id_22": datasets.Value("bool"),
		"soil_type_id_23": datasets.Value("bool"),
		"soil_type_id_24": datasets.Value("bool"),
		"soil_type_id_25": datasets.Value("bool"),
		"soil_type_id_26": datasets.Value("bool"),
		"soil_type_id_27": datasets.Value("bool"),
		"soil_type_id_28": datasets.Value("bool"),
		"soil_type_id_29": datasets.Value("bool"),
		"soil_type_id_30": datasets.Value("bool"),
		"soil_type_id_31": datasets.Value("bool"),
		"soil_type_id_32": datasets.Value("bool"),
		"soil_type_id_33": datasets.Value("bool"),
		"soil_type_id_34": datasets.Value("bool"),
		"soil_type_id_35": datasets.Value("bool"),
		"soil_type_id_36": datasets.Value("bool"),
		"soil_type_id_37": datasets.Value("bool"),
		"soil_type_id_38": datasets.Value("bool"),
		"soil_type_id_39": datasets.Value("bool"),
        "soil_type": datasets.Value("string"),
		"cover_type
    }
}

features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}


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


class Covertype(datasets.GeneratorBasedBuilder):
    # dataset versions
    DEFAULT_CONFIG = "covertype"
    BUILDER_CONFIGS = [
        CovertypeConfig(name="covertype",
                    description="Covertype for multiclass classification.")
    ]


    def _info(self):
        if self.config.name not in features_per_config:
            raise ValueError(f"Unknown configuration: {self.config.name}")
        
        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 = self.preprocess(data, config=self.config.name)

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

                yield row_id, data_row

    def preprocess(self, data: pandas.DataFrame, config: str = DEFAULT_CONFIG) -> pandas.DataFrame:
        data["500"] = data["500"].apply(lambda x: max(0, x)).astype(int)
        if "0.1" in data:
            data.drop("0.1", axis="columns", inplace=True)

        return data