import datasets
import pandas as pd

_CITATION = """\
@InProceedings{huggingface:dataset,
title = {botox-injections-before-and-after},
author = {TrainingDataPro},
year = {2023}
}
"""

_DESCRIPTION = """\
The dataset consists of photos featuring the same individuals captured before and after
botox injections procedure. The dataset contains a diverse range of individuals with
various ages, ethnicities and genders.
The dataset is useful for evaluation of the effectiveness of botox injections for
different skin and face types, face recognition and reidentification tasks. It can be
utilised for biometric tasks , in beauty sphere, for medical purposes and e-commerce.
"""
_NAME = "botox-injections-before-and-after"

_HOMEPAGE = f"https://huggingface.co/datasets/TrainingDataPro/{_NAME}"

_LICENSE = ""

_DATA = f"https://huggingface.co/datasets/TrainingDataPro/{_NAME}/resolve/main/data/"


class BotoxInjectionsBeforeAndAfter(datasets.GeneratorBasedBuilder):
    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("int32"),
                    "before": datasets.Image(),
                    "after": datasets.Image(),
                }
            ),
            supervised_keys=None,
            homepage=_HOMEPAGE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        before = dl_manager.download(f"{_DATA}before.tar.gz")
        after = dl_manager.download(f"{_DATA}after.tar.gz")
        annotations = dl_manager.download(f"{_DATA}{_NAME}.csv")
        before = dl_manager.iter_archive(before)
        after = dl_manager.iter_archive(after)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={
                    "before": before,
                    "after": after,
                    "annotations": annotations,
                },
            ),
        ]

    def _generate_examples(self, before, after, annotations):
        for idx, (
            (before_image_path, before_image),
            (after_image_path, after_image),
        ) in enumerate(zip(before, after)):
            yield idx, {
                "id": before_image_path.split("/")[-1].split(".")[0],
                "before": {"path": before_image_path, "bytes": before_image.read()},
                "after": {"path": after_image_path, "bytes": after_image.read()},
            }