Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -12,7 +12,7 @@ size_categories:
|
|
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/Kgt6S5S_zPGGQ3IlmyRVB.png" alt="Chart Example" style="width:
|
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.
|
|
|
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/Kgt6S5S_zPGGQ3IlmyRVB.png" alt="Chart Example" style="width: 800px; 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.
|