# Copyright 2020 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. # DocLayNet License: https://github.com/DS4SD/DocLayNet/blob/main/LICENSE # Apache License 2.0 """ DocLayNet small is a about 1% of the dataset DocLayNet (more information at https://huggingface.co/datasets/pierreguillou/DocLayNet-small) DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis DocLayNet dataset: - DocLayNet core dataset: https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_core.zip - DocLayNet extra dataset: https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_extra.zip """ import json import os import base64 from PIL import Image import datasets # Find for instance the citation on arxiv or on the dataset repo/website _CITATION = """\ @article{doclaynet2022, title = {DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis}, doi = {10.1145/3534678.353904}, url = {https://arxiv.org/abs/2206.01062}, author = {Pfitzmann, Birgit and Auer, Christoph and Dolfi, Michele and Nassar, Ahmed S and Staar, Peter W J}, year = {2022} } """ # You can copy an official description _DESCRIPTION = """\ Accurate document layout analysis is a key requirement for high-quality PDF document conversion. With the recent availability of public, large ground-truth datasets such as PubLayNet and DocBank, deep-learning models have proven to be very effective at layout detection and segmentation. While these datasets are of adequate size to train such models, they severely lack in layout variability since they are sourced from scientific article repositories such as PubMed and arXiv only. Consequently, the accuracy of the layout segmentation drops significantly when these models are applied on more challenging and diverse layouts. In this paper, we present \textit{DocLayNet}, a new, publicly available, document-layout annotation dataset in COCO format. It contains 80863 manually annotated pages from diverse data sources to represent a wide variability in layouts. For each PDF page, the layout annotations provide labelled bounding-boxes with a choice of 11 distinct classes. DocLayNet also provides a subset of double- and triple-annotated pages to determine the inter-annotator agreement. In multiple experiments, we provide smallline accuracy scores (in mAP) for a set of popular object detection models. We also demonstrate that these models fall approximately 10\% behind the inter-annotator agreement. Furthermore, we provide evidence that DocLayNet is of sufficient size. Lastly, we compare models trained on PubLayNet, DocBank and DocLayNet, showing that layout predictions of the DocLayNet-trained models are more robust and thus the preferred choice for general-purpose document-layout analysis. """ _HOMEPAGE = "https://developer.ibm.com/exchanges/data/all/doclaynet/" _LICENSE = "https://github.com/DS4SD/DocLayNet/blob/main/LICENSE" # The HuggingFace Datasets library doesn't host the datasets but only points to the original files. # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method) # _URLS = { # "first_domain": "https://huggingface.co/great-new-dataset-first_domain.zip", # "second_domain": "https://huggingface.co/great-new-dataset-second_domain.zip", # } # functions def load_image(image_path): image = Image.open(image_path).convert("RGB") w, h = image.size return image, (w, h) logger = datasets.logging.get_logger(__name__) class DocLayNetConfig(datasets.BuilderConfig): """BuilderConfig for DocLayNet small""" def __init__(self, **kwargs): """BuilderConfig for DocLayNet small. Args: **kwargs: keyword arguments forwarded to super. """ super(DocLayNetConfig, self).__init__(**kwargs) class DocLayNet(datasets.GeneratorBasedBuilder): """ DocLayNet small is a about 1% of the dataset DocLayNet (more information at https://huggingface.co/datasets/pierreguillou/DocLayNet-small) DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis DocLayNet dataset: - DocLayNet core dataset: https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_core.zip - DocLayNet extra dataset: https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_extra.zip """ VERSION = datasets.Version("1.1.0") # This is an example of a dataset with multiple configurations. # If you don't want/need to define several sub-sets in your dataset, # just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes. # If you need to make complex sub-parts in the datasets with configurable options # You can create your own builder configuration class to store attribute, inheriting from datasets.BuilderConfig # BUILDER_CONFIG_CLASS = MyBuilderConfig # You will be able to load one or the other configurations in the following list with # data = datasets.load_dataset('my_dataset', 'first_domain') # data = datasets.load_dataset('my_dataset', 'second_domain') BUILDER_CONFIGS = [ DocLayNetConfig(name="DocLayNet", version=datasets.Version("1.0.0"), description="DocLayNeT small dataset"), ] #DEFAULT_CONFIG_NAME = "DocLayNet" # It's not mandatory to have a default configuration. Just use one if it make sense. def _info(self): features = datasets.Features( { "id": datasets.Value("string"), "texts": datasets.Sequence(datasets.Value("string")), "bboxes_block": datasets.Sequence(datasets.Sequence(datasets.Value("int64"))), "bboxes_line": datasets.Sequence(datasets.Sequence(datasets.Value("int64"))), "categories": datasets.Sequence( datasets.features.ClassLabel( names=["Caption", "Footnote", "Formula", "List-item", "Page-footer", "Page-header", "Picture", "Section-header", "Table", "Text", "Title"] ) ), "image": datasets.features.Image(), "pdf": datasets.Value("string"), "page_hash": datasets.Value("string"), # unique identifier, equal to filename "original_filename": datasets.Value("string"), # original document filename "page_no": datasets.Value("int32"), # page number in original document "num_pages": datasets.Value("int32"), # total pages in original document "original_width": datasets.Value("int32"), # width in pixels @72 ppi "original_height": datasets.Value("int32"), # height in pixels @72 ppi "coco_width": datasets.Value("int32"), # with in pixels in PNG and COCO format "coco_height": datasets.Value("int32"), # with in pixels in PNG and COCO format "collection": datasets.Value("string"), # sub-collection name "doc_category": datasets.Value("string"), # category type of the document } ) return datasets.DatasetInfo( # This is the description that will appear on the datasets page. description=_DESCRIPTION, # This defines the different columns of the dataset and their types features=features, # Here we define them above because they are different between the two configurations # If there's a common (input, target) tuple from the features, uncomment supervised_keys line below and # specify them. They'll be used if as_supervised=True in builder.as_dataset. # supervised_keys=("sentence", "label"), # Homepage of the dataset for documentation homepage=_HOMEPAGE, # License for the dataset if available license=_LICENSE, # Citation for the dataset citation=_CITATION, ) def _split_generators(self, dl_manager): # TODO: This method is tasked with downloading/extracting the data and defining the splits depending on the configuration # If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name # dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLS # It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files. # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive downloaded_file = dl_manager.download_and_extract("https://huggingface.co/datasets/pierreguillou/DocLayNet-small/resolve/main/data/dataset_small.zip") splits = [] for split in self.config.splits: if split == "train": dataset = datasets.SplitGenerator( name=datasets.Split.TRAIN, # These kwargs will be passed to _generate_examples gen_kwargs={ "filepath": os.path.join(downloaded_file, "small_dataset/train/"), "split": "train", }, ) elif split in ["val", "valid", "validation", "dev"]: dataset = datasets.SplitGenerator( name=datasets.Split.VALIDATION, # These kwargs will be passed to _generate_examples gen_kwargs={ "filepath": os.path.join(downloaded_file, "small_dataset/val/"), "split": "val", }, ) elif split in ["test"]: dataset = datasets.SplitGenerator( name=datasets.Split.TEST, # These kwargs will be passed to _generate_examples gen_kwargs={ "filepath": os.path.join(downloaded_file, "small_dataset/test/"), "split": "test", }, ) else: continue splits.append(dataset) return splits def _generate_examples(self, filepath, split): logger.info("⏳ Generating examples from = %s", filepath) ann_dir = os.path.join(filepath, "annotations") img_dir = os.path.join(filepath, "images") pdf_dir = os.path.join(filepath, "pdfs") for guid, file in enumerate(sorted(os.listdir(ann_dir))): texts = [] bboxes_block = [] bboxes_line = [] categories = [] # get json file_path = os.path.join(ann_dir, file) with open(file_path, "r", encoding="utf8") as f: data = json.load(f) # get image image_path = os.path.join(img_dir, file) image_path = image_path.replace("json", "png") image, size = load_image(image_path) # get pdf pdf_path = os.path.join(pdf_dir, file) pdf_path = pdf_path.replace("json", "pdf") with open(pdf_path, "rb") as pdf_file: pdf_bytes = pdf_file.read() pdf_encoded_string = base64.b64encode(pdf_bytes) for item in data["form"]: text_example, category_example, bbox_block_example, bbox_line_example = item["text"], item["category"], item["box"], item["box_line"] texts.append(text_example) categories.append(category_example) bboxes_block.append(bbox_block_example) bboxes_line.append(bbox_line_example) # get all metadadata page_hash = data["metadata"]["page_hash"] original_filename = data["metadata"]["original_filename"] page_no = data["metadata"]["page_no"] num_pages = data["metadata"]["num_pages"] original_width = data["metadata"]["original_width"] original_height = data["metadata"]["original_height"] coco_width = data["metadata"]["coco_width"] coco_height = data["metadata"]["coco_height"] collection = data["metadata"]["collection"] doc_category = data["metadata"]["doc_category"] yield guid, {"id": str(guid), "texts": texts, "bboxes_block": bboxes_block, "bboxes_line": bboxes_line, "categories": categories, "image": image, "pdf": pdf_encoded_string, "page_hash": page_hash, "original_filename": original_filename, "page_no": page_no, "num_pages": num_pages, "original_width": original_width, "original_height": original_height, "coco_width": coco_width, "coco_height": coco_height, "collection": collection, "doc_category": doc_category}