titanic.py

"""Titanic"""

from typing import List

import datasets

import pandas


VERSION = datasets.Version("1.0.0")

DESCRIPTION = "Titanic dataset from the UCI ML repository."
_HOMEPAGE = "https://www.kaggle.com/datasets/vinicius150987/titanic3"
_URLS = ("https://www.kaggle.com/datasets/vinicius150987/titanic3")
_CITATION = """"""

# Dataset info
urls_per_split = {
	"train": "https://huggingface.co/datasets/mstz/titanic/raw/main/titanic.csv"
}
features_types_per_config = {
	"survival": {
		"passenger_class": datasets.Value("int8"),
		"is_male": datasets.Value("bool"),
		"age": datasets.Value("float64"),
		"sibsp": datasets.Value("float64"),
		"parch": datasets.Value("float64"),
		"ticket": datasets.Value("string"),
		"fare": datasets.Value("float64"),
		"cabin": datasets.Value("string"),
		"embarked": datasets.Value("string"),
		"has_survived": 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 TitanicConfig(datasets.BuilderConfig):
	def __init__(self, **kwargs):
		super(TitanicConfig, self).__init__(version=VERSION, **kwargs)
		self.features = features_per_config[kwargs["name"]]


class Titanic(datasets.GeneratorBasedBuilder):
	# dataset versions
	DEFAULT_CONFIG = "survival"
	BUILDER_CONFIGS = [
		TitanicConfig(name="survival",
					description="Titanic 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 = self.preprocess(data)

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

			yield row_id, data_row

	def preprocess(self, data):
		data = data.rename(columns={"sex": "is_male"})
		data = data[list(features_types_per_config[self.config.name].keys())]
		data.loc[:, "is_male"] = data.is_male.apply(lambda x: x == "male")
		data.loc[data.age == "?", "age"] = data.age.apply(lambda x: x if x != "?" else -1)

		return data