Datasets:
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cdla-permissive-2.0
|
3 |
+
task_categories:
|
4 |
+
- text-generation
|
5 |
+
tags:
|
6 |
+
- ocr
|
7 |
+
- chart
|
8 |
+
pretty_name: SynthChartNet
|
9 |
+
size_categories:
|
10 |
+
- 1M<n<10M
|
11 |
+
---
|
12 |
+
# SynthChartNet
|
13 |
+
|
14 |
+
<div style="display: flex; justify-content: center; align-items: center;">
|
15 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/663e1254887b6f5645a0399f/EhyekK2QdOe9PFID8PK4R.png" alt="Chart Example" style="width: 500px; height: auto">
|
16 |
+
</div>
|
17 |
+
|
18 |
+
**SynthChartNet** is a multimodal dataset designed for training the **SmolDocling** model on chart-based document understanding tasks. It consists of **1,981,157** synthetically generated samples, where each image depicts a chart (e.g., line chart, bar chart, pie chart, stacked bar chart), and the associated ground truth is given in **OTSL** format.
|
19 |
+
|
20 |
+
Charts were rendered using a diverse set of visualization libraries: **Matplotlib**, **Seaborn**, and **Pyecharts**, enabling visual variability in layout, style, and color schemes.
|
21 |
+
|
22 |
+
---
|
23 |
+
|
24 |
+
## Dataset Statistics
|
25 |
+
|
26 |
+
* **Total samples**: 1,981,157
|
27 |
+
|
28 |
+
* **Training set**: 1,981,157
|
29 |
+
|
30 |
+
* **Modalities**: Image, Text (OTSL format)
|
31 |
+
|
32 |
+
* **Chart Types**: Line, Bar, Pie, Stacked Bar
|
33 |
+
|
34 |
+
* **Rendering Engines**: Matplotlib, Seaborn, Pyecharts
|
35 |
+
|
36 |
+
---
|
37 |
+
|
38 |
+
## Data Format
|
39 |
+
|
40 |
+
Each dataset entry is structured as follows:
|
41 |
+
|
42 |
+
```json
|
43 |
+
{
|
44 |
+
"images": [PIL Image],
|
45 |
+
"texts": [
|
46 |
+
{
|
47 |
+
"assistant": "<loc_x0><loc_y0><loc_x1><loc_y1><_Chart_>OTSL_REPRESENTATION</chart>",
|
48 |
+
"source": "SynthChartNet",
|
49 |
+
"user": "<chart>"
|
50 |
+
}
|
51 |
+
]
|
52 |
+
}
|
53 |
+
```
|
54 |
+
|
55 |
+
---
|
56 |
+
|
57 |
+
## Intended Use
|
58 |
+
|
59 |
+
* Training multimodal models for **chart understanding**, specifically:
|
60 |
+
|
61 |
+
* Chart parsing and transcription to structured formats (OTSL)
|
62 |
+
|
63 |
+
---
|
64 |
+
|
65 |
+
## Citation
|
66 |
+
|
67 |
+
If you use SynthChartNet, please cite:
|
68 |
+
|
69 |
+
```bibtex
|
70 |
+
@article{nassar2025smoldocling,
|
71 |
+
title={SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion},
|
72 |
+
author={Nassar, Ahmed and Marafioti, Andres and Omenetti, Matteo and Lysak, Maksym and Livathinos, Nikolaos and Auer, Christoph and Morin, Lucas and de Lima, Rafael Teixeira and Kim, Yusik and Gurbuz, A Said and others},
|
73 |
+
journal={arXiv preprint arXiv:2503.11576},
|
74 |
+
year={2025}
|
75 |
+
}
|
76 |
+
```
|