p53.py

"""P53 Dataset"""

from typing import List
from functools import partial

import datasets

import pandas


VERSION = datasets.Version("1.0.0")

_ENCODING_DICS = {
	"class": {
		"inactive": 0,
		"active": 1
	}
}

DESCRIPTION = "P53 dataset."
_HOMEPAGE = "https://archive-beta.ics.uci.edu/dataset/170/p53"
_URLS = ("https://archive-beta.ics.uci.edu/dataset/170/p53")
_CITATION = """
@misc{misc_p53_mutants_188,
  author       = {Lathrop,Richard},
  title        = {{p53 Mutants}},
  year         = {2010},
  howpublished = {UCI Machine Learning Repository},
  note         = {{DOI}: \\url{10.24432/C5T89H}}
}
"""

# Dataset info
urls_per_split = {
	"train": "https://huggingface.co/datasets/mstz/p53/resolve/main/p53.data"
}
features_types_per_config = {
	"p53": {f"feature_{i}": datasets.Value("float64") for i in range(5408)}
}
features_types_per_config["p53"]["class"] = datasets.ClassLabel(num_classes=2)
features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}


class P53Config(datasets.BuilderConfig):
	def __init__(self, **kwargs):
		super(P53Config, self).__init__(version=VERSION, **kwargs)
		self.features = features_per_config[kwargs["name"]]


class P53(datasets.GeneratorBasedBuilder):
	# dataset versions
	DEFAULT_CONFIG = "p53"
	BUILDER_CONFIGS = [
		P53Config(name="p53", description="P53 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, header=None, nrows=100)
		data.columns = [f"feature_{i}" for i in range(5408)] + ["class"]
		print("preprocessing..")
		data = self.preprocess(data)
		print("preprocessed!\n\n\n\n")

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

			yield row_id, data_row

	def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
		for feature in _ENCODING_DICS:
			print(f"encoding {feature}\n\n\n")
			encoding_function = partial(self.encode, feature)
			data.loc[:, feature] = data[feature].apply(encoding_function)
		
		for feature in data.columns:
			if feature == "class":
				break
			data.loc[data[feature] == "?", feature] = data[data[feature] != "?"].astype(float).mean()

		return data[list(features_types_per_config[self.config.name].keys())]

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