davanstrien HF Staff commited on
Commit
1121c4d
·
verified ·
1 Parent(s): 7137397

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +8 -39
README.md CHANGED
@@ -7,37 +7,6 @@ tags:
7
  - markdown
8
  - uv-script
9
  - generated
10
- dataset_info:
11
- features:
12
- - name: document_id
13
- dtype: string
14
- - name: page_number
15
- dtype: string
16
- - name: image
17
- dtype: image
18
- - name: text
19
- dtype: string
20
- - name: alto_xml
21
- dtype: string
22
- - name: has_image
23
- dtype: bool
24
- - name: has_alto
25
- dtype: bool
26
- - name: layout
27
- dtype: string
28
- - name: inference_info
29
- dtype: string
30
- splits:
31
- - name: train
32
- num_bytes: 440695404
33
- num_examples: 2000
34
- download_size: 326516177
35
- dataset_size: 440695404
36
- configs:
37
- - config_name: default
38
- data_files:
39
- - split: train
40
- path: data/train-*
41
  ---
42
 
43
  # Document OCR using dots.ocr
@@ -49,8 +18,8 @@ This dataset contains OCR results from images in [NationalLibraryOfScotland/Brit
49
  - **Source Dataset**: [NationalLibraryOfScotland/Britain-and-UK-Handbooks-Dataset](https://huggingface.co/datasets/NationalLibraryOfScotland/Britain-and-UK-Handbooks-Dataset)
50
  - **Model**: [rednote-hilab/dots.ocr](https://huggingface.co/rednote-hilab/dots.ocr)
51
  - **Number of Samples**: 2,000
52
- - **Processing Time**: 19.2 min
53
- - **Processing Date**: 2025-10-09 13:22 UTC
54
 
55
  ### Configuration
56
 
@@ -58,9 +27,9 @@ This dataset contains OCR results from images in [NationalLibraryOfScotland/Brit
58
  - **Output Column**: `markdown`
59
  - **Dataset Split**: `train`
60
  - **Batch Size**: 512
61
- - **Prompt Mode**: layout-all
62
- - **Max Model Length**: 16,200 tokens
63
- - **Max Output Tokens**: 16,000
64
  - **GPU Memory Utilization**: 80.0%
65
 
66
  ## Model Information
@@ -108,9 +77,9 @@ uv run https://huggingface.co/datasets/uv-scripts/ocr/raw/main/dots-ocr.py \
108
  <output-dataset> \
109
  --image-column image \
110
  --batch-size 512 \
111
- --prompt-mode layout-all \
112
- --max-model-len 16200 \
113
- --max-tokens 16000 \
114
  --gpu-memory-utilization 0.8
115
  ```
116
 
 
7
  - markdown
8
  - uv-script
9
  - generated
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10
  ---
11
 
12
  # Document OCR using dots.ocr
 
18
  - **Source Dataset**: [NationalLibraryOfScotland/Britain-and-UK-Handbooks-Dataset](https://huggingface.co/datasets/NationalLibraryOfScotland/Britain-and-UK-Handbooks-Dataset)
19
  - **Model**: [rednote-hilab/dots.ocr](https://huggingface.co/rednote-hilab/dots.ocr)
20
  - **Number of Samples**: 2,000
21
+ - **Processing Time**: 12.5 min
22
+ - **Processing Date**: 2025-10-09 13:45 UTC
23
 
24
  ### Configuration
25
 
 
27
  - **Output Column**: `markdown`
28
  - **Dataset Split**: `train`
29
  - **Batch Size**: 512
30
+ - **Prompt Mode**: layout-only
31
+ - **Max Model Length**: 8,192 tokens
32
+ - **Max Output Tokens**: 8,192
33
  - **GPU Memory Utilization**: 80.0%
34
 
35
  ## Model Information
 
77
  <output-dataset> \
78
  --image-column image \
79
  --batch-size 512 \
80
+ --prompt-mode layout-only \
81
+ --max-model-len 8192 \
82
+ --max-tokens 8192 \
83
  --gpu-memory-utilization 0.8
84
  ```
85