DocumentFigureClassifier v2.0 EU AI Act Compliance Profile

EfficientNet-B0 Document Image Classifier

This is an image classification model based on Google EfficientNet-B0, fine-tuned to classify input images into one of the following 26 categories:

1. logo
2. photograph
3. icon
4. engineering_drawing
5. line_chart
6. bar_chart
7. other
8. table
9. flow_chart
10. screenshot_from_computer
11. signature
12. screenshot_from_manual
13. geographical_map
14. pie_chart
15. page_thumbnail
16. stamp
17. music
18. calendar
19. qr_code
20. bar_code
21. full_page_image
22. scatter_plot
23. chemistry_structure
24. topographical_map
25. crossword_puzzle
26. box_plot

How to use - Transformers

Example of how to classify an image into one of the 26 classes using transformers:

import torch
import torchvision.transforms as transforms

from transformers import EfficientNetForImageClassification
from PIL import Image
import requests


urls = [
    'http://images.cocodataset.org/val2017/000000039769.jpg',
    'http://images.cocodataset.org/test-stuff2017/000000001750.jpg',
    'http://images.cocodataset.org/test-stuff2017/000000000001.jpg'
]

image_processor = transforms.Compose(
    [
        transforms.Resize((224, 224)),
        transforms.ToTensor(),
        transforms.Normalize(
            mean=[0.485, 0.456, 0.406],
            std=[0.47853944, 0.4732864, 0.47434163],
        ),
    ]
)

images = []
for url in urls:
    image = Image.open(requests.get(url, stream=True).raw).convert("RGB")
    image = image_processor(image)
    images.append(image)


model_id = 'docling-project/DocumentFigureClassifier-v2.0'

model = EfficientNetForImageClassification.from_pretrained(model_id)

labels = model.config.id2label

device = torch.device("cpu")

torch_images = torch.stack(images).to(device)

with torch.no_grad():
    logits = model(torch_images).logits  # (batch_size, num_classes)
    probs_batch = logits.softmax(dim=1)

Your obligations as a DocumentFigureClassifier v2.0 deployer

Track your progress

Sign up to track obligations, receive deadline reminders, and generate audit packages.

Tags

Base Model