cache_loader.py

import os
from io import BytesIO
from tqdm import tqdm
from typing import Callable, Dict, List, Literal, NamedTuple, Optional, Tuple, Union

from datasets import load_dataset, logging
from PIL import Image

import fitz as PyMuPDF

logger = logging.get_logger(__name__)

def pymupdf_image_extraction(data: bytes):
    images = []
    with PyMuPDF.open(stream=data, filetype="pdf") as pdf_file:
        for page_index in range(len(pdf_file)):
            page = pdf_file[page_index]
            for image_index, img in enumerate(page.get_images(), start=1):
                xref = img[0]
                base_image = pdf_file.extract_image(xref)
                image_bytes = base_image["image"]
                im = Image.open(BytesIO(image_bytes))
                images.append(im) #<PIL.PngImagePlugin.PngImageFile image mode=L size=2560x3320 at 0x7FD32E7430A0>
    return images


def batch_get_images(streams):
    return {
        "images": 
            [pymupdf_image_extraction(pdf_stream) for pdf_stream in tqdm(streams,  desc="get_images")]
        
    }


def pymupdf_page_extraction(pdf_stream):
    try:
        return len(PyMuPDF.open(stream=pdf_stream, filetype="pdf"))
    except Exception as e:
        logger.warning(f"{e}")
        return 0


def batch_get_pages(streams):
    return {
        "pages": 
            [pymupdf_page_extraction(pdf_stream) for pdf_stream in tqdm(streams,  desc="get_pages")],
    }


testds = load_dataset("jordyvl/rvl_cdip_multi", cache_dir="/mnt/lerna/data/HFcache", split="test", revision="d3a654c9f63f14d0aaa94e08aa30aa3dc20713c1")

func = lambda batch: batch_get_pages(batch["file"])
testds = testds.map(func, batched=True, keep_in_memory=False)
testds = testds.filter(lambda example: example["pages"] != 0)

func = lambda batch: batch_get_images(batch["file"])
testds = testds.map(func, batched=True, keep_in_memory=False)

import pdb; pdb.set_trace()  # breakpoint 70086661 //

"""
Filter based on page count?
"""