sroie2019_sample.py

import torch
import datasets

from pathlib import Path
from torch.utils.data import Dataset

from datasets import load_dataset, Features, Value, ClassLabel, DownloadConfig

_DESCRIPTION = """\
"""

_CITATION = """\
"""

# it could be file or url path
_TRAIN_DOWNLOAD_URL = "train.txt"
_VAL_DOWNLOAD_URL = "val.txt"

CLASS_NAMES = ["company", "date", "address", "total", "O"]


class CustomTokenDataset(datasets.GeneratorBasedBuilder):
    """CustomTokenDataset dataset."""

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "tokens": datasets.Sequence(datasets.Value("string")),
                    "ner_tags": datasets.Sequence(
                        datasets.features.ClassLabel(names=sorted(list(CLASS_NAMES)))
                    ),
                }
            ),
            supervised_keys=None,
            homepage="",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        urls_to_download = {
            "train": _TRAIN_DOWNLOAD_URL,
            "val": _VAL_DOWNLOAD_URL,
        }
        downloaded_files = dl_manager.download_and_extract(urls_to_download)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filepath": downloaded_files["train"]},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                gen_kwargs={"filepath": downloaded_files["val"]},
            ),
        ]

    def _generate_examples(self, filepath):
        with open(filepath, encoding="utf-8") as f:
            guid = 0
            tokens = []
            ner_tags = []
            for line in f:
                if line == "" or line == "\n":
                    if tokens:
                        yield guid, {
                            "id": str(guid),
                            "tokens": tokens,
                            "ner_tags": ner_tags,
                        }
                        guid += 1
                        tokens = []
                        ner_tags = []
                else:
                    # CustomDataset tokens are space separated
                    splits = line.split(" ")
                    tokens.append(splits[0])
                    ner_tags.append(splits[1].rstrip())
            # last example
            yield guid, {
                "id": str(guid),
                "tokens": tokens,
                "ner_tags": ner_tags,
            }