Object Detection
PaddleOCR
Safetensors
English
Chinese
multilingual
pp_doclayout_v2
PaddlePaddle
ocr
layout
layout_detection
Instructions to use PaddlePaddle/PP-DocLayoutV2_safetensors with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PaddleOCR
How to use PaddlePaddle/PP-DocLayoutV2_safetensors with PaddleOCR:
# 1. See https://www.paddlepaddle.org.cn/en/install to install paddlepaddle # 2. pip install paddleocr from paddleocr import LayoutDetection model = LayoutDetection(model_name="PP-DocLayoutV2_safetensors") output = model.predict(input="path/to/image.png", batch_size=1) for res in output: res.print() res.save_to_img(save_path="./output/") res.save_to_json(save_path="./output/res.json") - Notebooks
- Google Colab
- Kaggle
Update config.json
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -128,7 +128,7 @@
|
|
| 128 |
"shape_size": 170,
|
| 129 |
"num_classes": 20,
|
| 130 |
"relation_bias_embed_dim": 16,
|
| 131 |
-
"
|
| 132 |
"relation_bias_scale": 100,
|
| 133 |
"global_pointer_head_size": 64,
|
| 134 |
"tril_mask": true
|
|
|
|
| 128 |
"shape_size": 170,
|
| 129 |
"num_classes": 20,
|
| 130 |
"relation_bias_embed_dim": 16,
|
| 131 |
+
"relation_bias_theta": 10000,
|
| 132 |
"relation_bias_scale": 100,
|
| 133 |
"global_pointer_head_size": 64,
|
| 134 |
"tril_mask": true
|