"""Ozone: A Census Dataset""" from typing import List import datasets import pandas VERSION = datasets.Version("1.0.0") _BASE_FEATURE_NAMES = [ "Date", "WSR0", "WSR1", "WSR2", "WSR3", "WSR4", "WSR5", "WSR6", "WSR7", "WSR8", "WSR9", "WSR10", "WSR11", "WSR12", "WSR13", "WSR14", "WSR15", "WSR16", "WSR17", "WSR18", "WSR19", "WSR20", "WSR21", "WSR22", "WSR23", "WSR_PK", "WSR_AV", "T0", "T1", "T2", "T3", "T4", "T5", "T6", "T7", "T8", "T9", "T10", "T11", "T12", "T13", "T14", "T15", "T16", "T17", "T18", "T19", "T20", "T21", "T22", "T23", "T_PK", "T_AV", "T85", "RH85", "U85", "V85", "HT85", "T70", "RH70", "U70", "V70", "HT70", "T50", "RH50", "U50", "V50", "HT50", "KI", "TT", "SLP", "SLP_", "Precp", "Class" ] DESCRIPTION = "Ozone dataset from the UCI ML repository." _HOMEPAGE = "https://archive.ics.uci.edu/ml/datasets/Ozone" _URLS = ("https://archive.ics.uci.edu/ml/datasets/Ozone") _CITATION = """ @misc{misc_ozone_level_detection_172, author = {Zhang,Kun, Fan,Wei & Yuan,XiaoJing}, title = {{Ozone Level Detection}}, year = {2008}, howpublished = {UCI Machine Learning Repository}, note = {{DOI}: \\url{10.24432/C5NG6W}} }""" # Dataset info urls_per_split = { "8hr": {"train": "https://huggingface.co./datasets/mstz/ozone/raw/main/eighthr.data"}, "1hr": {"train": "https://huggingface.co./datasets/mstz/ozone/raw/main/onehr.data"}, } features_types_per_config = { "8hr": { "WSR0": datasets.Value("float64"), "WSR1": datasets.Value("float64"), "WSR2": datasets.Value("float64"), "WSR3": datasets.Value("float64"), "WSR4": datasets.Value("float64"), "WSR5": datasets.Value("float64"), "WSR6": datasets.Value("float64"), "WSR7": datasets.Value("float64"), "WSR8": datasets.Value("float64"), "WSR9": datasets.Value("float64"), "WSR10": datasets.Value("float64"), "WSR11": datasets.Value("float64"), "WSR12": datasets.Value("float64"), "WSR13": datasets.Value("float64"), "WSR14": datasets.Value("float64"), "WSR15": datasets.Value("float64"), "WSR16": datasets.Value("float64"), "WSR17": datasets.Value("float64"), "WSR18": datasets.Value("float64"), "WSR19": datasets.Value("float64"), "WSR20": datasets.Value("float64"), "WSR21": datasets.Value("float64"), "WSR22": datasets.Value("float64"), "WSR23": datasets.Value("float64"), "WSR_PK": datasets.Value("float64"), "WSR_AV": datasets.Value("float64"), "T0": datasets.Value("float64"), "T1": datasets.Value("float64"), "T2": datasets.Value("float64"), "T3": datasets.Value("float64"), "T4": datasets.Value("float64"), "T5": datasets.Value("float64"), "T6": datasets.Value("float64"), "T7": datasets.Value("float64"), "T8": datasets.Value("float64"), "T9": datasets.Value("float64"), "T10": datasets.Value("float64"), "T11": datasets.Value("float64"), "T12": datasets.Value("float64"), "T13": datasets.Value("float64"), "T14": datasets.Value("float64"), "T15": datasets.Value("float64"), "T16": datasets.Value("float64"), "T17": datasets.Value("float64"), "T18": datasets.Value("float64"), "T19": datasets.Value("float64"), "T20": datasets.Value("float64"), "T21": datasets.Value("float64"), "T22": datasets.Value("float64"), "T23": datasets.Value("float64"), "T_PK": datasets.Value("float64"), "T_AV": datasets.Value("float64"), "T85": datasets.Value("float64"), "RH85": datasets.Value("float64"), "U85": datasets.Value("float64"), "V85": datasets.Value("float64"), "HT85": datasets.Value("float64"), "T70": datasets.Value("float64"), "RH70": datasets.Value("float64"), "U70": datasets.Value("float64"), "V70": datasets.Value("float64"), "HT70": datasets.Value("float64"), "T50": datasets.Value("float64"), "RH50": datasets.Value("float64"), "U50": datasets.Value("float64"), "V50": datasets.Value("float64"), "HT50": datasets.Value("float64"), "KI": datasets.Value("float64"), "TT": datasets.Value("float64"), "SLP": datasets.Value("float64"), "SLP_": datasets.Value("float64"), "Precp": datasets.Value("float64"), "Class": datasets.ClassLabel(num_classes=2, names=("no", "yes")) }, "1hr": { "WSR0": datasets.Value("float64"), "WSR1": datasets.Value("float64"), "WSR2": datasets.Value("float64"), "WSR3": datasets.Value("float64"), "WSR4": datasets.Value("float64"), "WSR5": datasets.Value("float64"), "WSR6": datasets.Value("float64"), "WSR7": datasets.Value("float64"), "WSR8": datasets.Value("float64"), "WSR9": datasets.Value("float64"), "WSR10": datasets.Value("float64"), "WSR11": datasets.Value("float64"), "WSR12": datasets.Value("float64"), "WSR13": datasets.Value("float64"), "WSR14": datasets.Value("float64"), "WSR15": datasets.Value("float64"), "WSR16": datasets.Value("float64"), "WSR17": datasets.Value("float64"), "WSR18": datasets.Value("float64"), "WSR19": datasets.Value("float64"), "WSR20": datasets.Value("float64"), "WSR21": datasets.Value("float64"), "WSR22": datasets.Value("float64"), "WSR23": datasets.Value("float64"), "WSR_PK": datasets.Value("float64"), "WSR_AV": datasets.Value("float64"), "T0": datasets.Value("float64"), "T1": datasets.Value("float64"), "T2": datasets.Value("float64"), "T3": datasets.Value("float64"), "T4": datasets.Value("float64"), "T5": datasets.Value("float64"), "T6": datasets.Value("float64"), "T7": datasets.Value("float64"), "T8": datasets.Value("float64"), "T9": datasets.Value("float64"), "T10": datasets.Value("float64"), "T11": datasets.Value("float64"), "T12": datasets.Value("float64"), "T13": datasets.Value("float64"), "T14": datasets.Value("float64"), "T15": datasets.Value("float64"), "T16": datasets.Value("float64"), "T17": datasets.Value("float64"), "T18": datasets.Value("float64"), "T19": datasets.Value("float64"), "T20": datasets.Value("float64"), "T21": datasets.Value("float64"), "T22": datasets.Value("float64"), "T23": datasets.Value("float64"), "T_PK": datasets.Value("float64"), "T_AV": datasets.Value("float64"), "T85": datasets.Value("float64"), "RH85": datasets.Value("float64"), "U85": datasets.Value("float64"), "V85": datasets.Value("float64"), "HT85": datasets.Value("float64"), "T70": datasets.Value("float64"), "RH70": datasets.Value("float64"), "U70": datasets.Value("float64"), "V70": datasets.Value("float64"), "HT70": datasets.Value("float64"), "T50": datasets.Value("float64"), "RH50": datasets.Value("float64"), "U50": datasets.Value("float64"), "V50": datasets.Value("float64"), "HT50": datasets.Value("float64"), "KI": datasets.Value("float64"), "TT": datasets.Value("float64"), "SLP": datasets.Value("float64"), "SLP_": datasets.Value("float64"), "Precp": datasets.Value("float64"), "Class": 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 OzoneConfig(datasets.BuilderConfig): def __init__(self, **kwargs): super(OzoneConfig, self).__init__(version=VERSION, **kwargs) self.features = features_per_config[kwargs["name"]] class Ozone(datasets.GeneratorBasedBuilder): # dataset versions DEFAULT_CONFIG = "8hr" BUILDER_CONFIGS = [ OzoneConfig(name="8hr", description="Ozone for binary classification."), OzoneConfig(name="1hr", description="Ozone 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[self.config.name]["train"]}) ] def _generate_examples(self, filepath: str): data = pandas.read_csv(filepath) data.drop("Date", axis="columns", inplace=True) data.loc[:, "Class"] = data.Class.astype(int) data = data[~(data.isin(["?"]).any(axis=1))] data = data.infer_objects() for row_id, row in data.iterrows(): data_row = dict(row) yield row_id, data_row