Sponsored by
APIMart
ufal / vit-historical-page
Model's Last Updated: July 23 2026
image-classification
Introduction of vit-historical-page
Model Details of vit-historical-page
Image classification using fine-tuned ViT - for historical :bowtie: documents sorting
Goal: solve a task of archive page images sorting (for their further content-based processing)
Scope: Processing of images, training and evaluation of ViT model, input file/directory processing, class ๐ท๏ธ (category) results of top N predictions output, predictions summarizing into a tabular format, HF ๐ hub support for the model
Model description ๐
๐ฒ Fine-tuned model repository: vit-historical-page ^1 ๐
๐ณ Base model repository: google's vit-base-patch16-224 ^2 ๐
Data ๐
Training set of the model: 8950 images
Categories ๐ท๏ธ
| Label๏ธ | Ratio | Description |
|---|---|---|
| DRAW | 11.89% | ๐ - drawings, maps, paintings with text |
| DRAW_L | 8.17% | ๐๐ - drawings ... with a table legend or inside tabular layout / forms |
| LINE_HW | 5.99% | โ๏ธ๐ - handwritten text lines inside tabular layout / forms |
| LINE_P | 6.06% | ๐ - printed text lines inside tabular layout / forms |
| LINE_T | 13.39% | ๐ - machine typed text lines inside tabular layout / forms |
| PHOTO | 10.21% | ๐ - photos with text |
| PHOTO_L | 7.86% | ๐๐ - photos inside tabular layout / forms or with a tabular annotation |
| TEXT | 8.58% | ๐ฐ - mixed types of printed and handwritten texts |
| TEXT_HW | 7.36% | โ๏ธ๐ - only handwritten text |
| TEXT_P | 6.95% | ๐ - only printed text |
| TEXT_T | 13.53% | ๐ - only machine typed text |
Evaluation set (same proportions): 995 images
Data preprocessing
During training the following transforms were applied randomly with a 50% chance:
- transforms.ColorJitter(brightness 0.5)
- transforms.ColorJitter(contrast 0.5)
- transforms.ColorJitter(saturation 0.5)
- transforms.ColorJitter(hue 0.5)
- transforms.Lambda(lambda img: ImageEnhance.Sharpness(img).enhance(random.uniform(0.5, 1.5)))
- transforms.Lambda(lambda img: img.filter(ImageFilter.GaussianBlur(radius=random.uniform(0, 2))))
Training Hyperparameters
- eval_strategy "epoch"
- save_strategy "epoch"
- learning_rate 5e-5
- per_device_train_batch_size 8
- per_device_eval_batch_size 8
- num_train_epochs 3
- warmup_ratio 0.1
- logging_steps 10
- load_best_model_at_end True
- metric_for_best_model "accuracy"
Results ๐
Evaluation set's accuracy ( Top-3 ): 99.6%
Evaluation set's accuracy ( Top-1 ): 97.3%
Result tables
-
Manually โ checked evaluation dataset results (TOP-3): model_TOP-3_EVAL.csv ๐
-
Manually โ checked evaluation dataset results (TOP-1): model_TOP-1_EVAL.csv ๐
Table columns
- FILE - name of the file
- PAGE - number of the page
- CLASS-N - label of the category ๐ท๏ธ, guess TOP-N
- SCORE-N - score of the category ๐ท๏ธ, guess TOP-N
- TRUE - actual label of the category ๐ท๏ธ
Contacts ๐ง
For support write to ๐ง [email protected] ๐ง
Official repository: UFAL ^3
Acknowledgements ๐
- Developed by UFAL ^5 ๐ฅ
- Funded by ATRIUM ^4 ๐ฐ
- Shared by ATRIUM ^4 & UFAL ^5
- Model type: fine-tuned ViT ^2 with a 224x224 resolution size
ยฉ๏ธ 2022 UFAL & ATRIUM
Runs of ufal vit-historical-page on huggingface.co
2.4K
Total runs
0
24-hour runs
0
3-day runs
20
7-day runs
2.2K
30-day runs
More Information About vit-historical-page huggingface.co Model
Url of vit-historical-page
vit-historical-page huggingface.co Url
Provider of vit-historical-page huggingface.co
Other API from ufal
Total runs:
107
Run Growth:
-16
Growth Rate:
-14.95%
Total runs:
60
Run Growth:
32
Growth Rate:
53.33%
Total runs:
57
Run Growth:
27
Growth Rate:
47.37%
Total runs:
39
Run Growth:
-28
Growth Rate:
-71.79%
Total runs:
34
Run Growth:
21
Growth Rate:
60.00%
Total runs:
31
Run Growth:
-6
Growth Rate:
-19.35%
Total runs:
27
Run Growth:
3
Growth Rate:
11.11%
Total runs:
23
Run Growth:
-3
Growth Rate:
-13.04%
Total runs:
17
Run Growth:
-1
Growth Rate:
-5.88%
Total runs:
17
Run Growth:
-4
Growth Rate:
-23.53%
Total runs:
17
Run Growth:
-8
Growth Rate:
-47.06%
Total runs:
14
Run Growth:
-9
Growth Rate:
-64.29%
Total runs:
14
Run Growth:
-6
Growth Rate:
-42.86%
Total runs:
12
Run Growth:
-25
Growth Rate:
-208.33%
Total runs:
10
Run Growth:
-5
Growth Rate:
-50.00%
Total runs:
10
Run Growth:
-16
Growth Rate:
-160.00%
Total runs:
9
Run Growth:
-15
Growth Rate:
-166.67%
Total runs:
9
Run Growth:
5
Growth Rate:
55.56%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%