iris.py

from typing import List
from functools import partial

import datasets

import pandas


VERSION = datasets.Version("1.0.0")


DESCRIPTION = "Iris efficiency dataset from the UCI repository."
_HOMEPAGE = "https://archive-beta.ics.uci.edu/dataset/53/iris"
_URLS = ("https://archive-beta.ics.uci.edu/dataset/53/iris")
_CITATION = """
@misc{misc_iris_53,
  author       = {Fisher,R. A. & Fisher,R.A.},
  title        = {{Iris}},
  year         = {1988},
  howpublished = {UCI Machine Learning Repository},
  note         = {{DOI}: \\url{10.24432/C56C76}}
}"""

# Dataset info
_BASE_FEATURE_NAMES = [
	"sepal_length",
	"sepal_width",
	"petal_length",
	"petal_width",
	"class"
]
urls_per_split = {
	"train": "https://huggingface.co/datasets/mstz/iris/raw/main/iris.data"
}
features_types_per_config = {
	"iris": {
		"sepal_length": datasets.Value("float64"),
		"sepal_width": datasets.Value("float64"),
		"petal_length": datasets.Value("float64"),
		"petal_width": datasets.Value("float64"),
		"class": datasets.ClassLabel(num_classes=3, names=("setosa", "versicolor", "virginica"))
	},
	"setosa": {
		"sepal_length": datasets.Value("float64"),
		"sepal_width": datasets.Value("float64"),
		"petal_length": datasets.Value("float64"),
		"petal_width": datasets.Value("float64"),
		"class": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
	},
	"versicolor": {
		"sepal_length": datasets.Value("float64"),
		"sepal_width": datasets.Value("float64"),
		"petal_length": datasets.Value("float64"),
		"petal_width": datasets.Value("float64"),
		"class": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
	},
	"virginica": {
		"sepal_length": datasets.Value("float64"),
		"sepal_width": datasets.Value("float64"),
		"petal_length": datasets.Value("float64"),
		"petal_width": 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 IrisConfig(datasets.BuilderConfig):
	def __init__(self, **kwargs):
		super(IrisConfig, self).__init__(version=VERSION, **kwargs)
		self.features = features_per_config[kwargs["name"]]


class Iris(datasets.GeneratorBasedBuilder):
	# dataset versions
	DEFAULT_CONFIG = "iris"
	BUILDER_CONFIGS = [
		IrisConfig(name="iris", description="Iris dataset."),
		IrisConfig(name="setosa", description="Binary classification of setosa."),
		IrisConfig(name="versicolor", description="Binary classification of versicolor."),
		IrisConfig(name="virginica", description="Binary classification of virginica.")		
	]


	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, header=None)
		data = self.preprocess(data)

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

			yield row_id, data_row
	
	def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
		data.columns = _BASE_FEATURE_NAMES
		data.loc[:, "class"] = data["class"].apply(lambda x: {
			"Iris-setosa": 0,
			"Iris-versicolor": 1,
			"Iris-virginica": 2
		}[x])
		

		if self.config.name == "setosa":
			data.loc[:, "class"] = data["class"].apply(lambda x: 1 if x == 0 else 0)
		elif self.config.name == "versicolor":
			data.loc[:, "class"] = data["class"].apply(lambda x: 1 if x == 1 else 0)
		if self.config.name == "virginica":
			data.loc[:, "class"] = data["class"].apply(lambda x: 1 if x == 2 else 0)

		return data