rvl_cdip_mp.py

# Copyright 2023 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""RVL-CDIP_mp (Ryerson Vision Lab Complex Document Information Processing) -Extended -Multipage dataset"""


import os
import datasets
from pathlib import Path
from typing import List
from tqdm import tqdm

datasets.logging.set_verbosity_info()
logger = datasets.logging.get_logger(__name__)

MODE = "binary"

_CITATION = """
    
@inproceedings{bdpc,
    title = {Beyond Document Page Classification},
    author = {Anonymous},
    booktitle = {Under Review},
    year = {2023}
}
"""

_DESCRIPTION = """\
The RVL-CDIP (Ryerson Vision Lab Complex Document Information Processing) dataset consists of originally retrieved documents in 16 classes.
There were +-500 documents from the original dataset that could not be retrieved based on the metadata or were corrupt in IDL. 
"""


_HOMEPAGE = "https://www.cs.cmu.edu/~aharley/rvl-cdip/"
_LICENSE = "https://www.industrydocuments.ucsf.edu/help/copyright/"


SOURCE = "bdpc/rvl_cdip_mp"

_BACKOFF_folder = "/mnt/lerna/data/RVL-CDIP_pdf"

_CLASSES = [
    "letter",
    "form",
    "email",
    "handwritten",
    "advertisement",
    "scientific_report",
    "scientific_publication",
    "specification",
    "file_folder",
    "news_article",
    "budget",
    "invoice",
    "presentation",
    "questionnaire",
    "resume",
    "memo",
]


def open_pdf_binary(pdf_file):
    with open(pdf_file, "rb") as f:
        return f.read()


class RvlCdipMp(datasets.GeneratorBasedBuilder):
    BUILDER_CONFIGS = [
        datasets.BuilderConfig(
            name="default",
            version=datasets.Version("1.0.0", ""),
            description="",
        )
    ]

    def __init__(self, *args, examples_per_class=None, **kwargs):
        super().__init__(*args, **kwargs)
        # examples per class to stop generating
        self.examples_per_class = examples_per_class

    @property
    def manual_download_instructions(self):
        return (
            "To use RVL-CDIP_multi you have to download it manually. Please extract all files in one folder and load the dataset with: "
            "`datasets.load_dataset('bdpc/rvl_cdip_mp', data_dir='path/to/folder/folder_name')`"
        )

    def _info(self):
        # DEFAULT_WRITER_BATCH_SIZE

        folder = None
        if isinstance(self.config.data_files, str):
            folder = self.config.data_files  # needs to be extracted cuz zip/tar
        else:
            if isinstance(self.config.data_dir, str):
                folder = self.config.data_dir  # contains the folder structure at someone local disk
            else:
                folder = _BACKOFF_folder  # my local path, others should set data_dir or data_files
            self.config.data_dir = folder

        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "file": datasets.Value("binary"),  # datasets.Sequence(datasets.Image()),
                    "labels": datasets.features.ClassLabel(names=_CLASSES),
                }
            ),
            task_templates=None,
        )

    def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
        if os.path.isdir(self.config.data_dir):
            data_files = {
                labelset: os.path.join(self.config.data_dir, labelset)
                for labelset in sorted(os.listdir(self.config.data_dir), reverse=True)
                if not "csv" in labelset
            }

        elif self.config.data_dir.endswith(".tar.gz"):
            archive_path = dl_manager.download(self.config.data_dir)
            data_files = dl_manager.iter_archive(archive_path)
            raise NotImplementedError()
        elif self.config.data_dir.endswith(".zip"):
            archive_path = dl_manager.download_and_extract(self.config.data_dir)
            data_files = dl_manager.iter_archive(archive_path)
            raise NotImplementedError()

        splits = []
        for split_name, folder in data_files.items():
            print(folder)
            splits.append(datasets.SplitGenerator(name=split_name, gen_kwargs={"archive_path": folder}))
        return splits

    def _generate_examples(self, archive_path):
        labels = self.info.features["labels"]

        extensions = {".pdf", ".PDF"}

        for i, path in tqdm(enumerate(Path(archive_path).glob("**/*/*")), desc=f"{archive_path}"):
            if path.suffix in extensions:
                try:
                    images = open_pdf_binary(path)
                    yield path.name, {
                        "file": images,
                        "labels": labels.encode_example(path.parent.name.lower()),
                    }
                except Exception as e:
                    logger.warning(f"{e} failed to parse {i}")