Replies: 1 comment
-
如果出现表格结构异常的话,大概率还是因为模型在您的场景中的效果不好导致的,可能需要继续训练。当然关于表格识别本身,我们后续会release更好的模型,敬请期待。 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
单独训练了文字检测、识别模块、表格结构识别模块(SLANet_plus),现在文字检测和识别都正常,但是表格结构异常


如上图所示,文字检测和识别正常,但是转换表格结构异常(已针对此格式表格单独训练了SLANet_plus),
这种情况应该如何处理?
代码使用方法如下:
pipeline = PPStructureV3(
# 文本检测
text_detection_model_name="PP-OCRv5_server_det",
text_detection_model_dir="G:\Python\YOLOv11\PP-OCRv5_server_det_infer3",
text_det_unclip_ratio=2.0,
text_det_limit_type="max",
text_det_limit_side_len=1080,
# 文字识别
text_recognition_model_name='PP-OCRv5_server_rec',
text_recognition_model_dir='G:\Python\YOLOv11\PP-OCRv5_server_rec_infer',
# 有线表格结构识别
wired_table_structure_recognition_model_name="SLANet_plus",
wired_table_structure_recognition_model_dir="G:\Python\YOLOv11\SLANet_finetune_infer1",
# 无线表格结构识别
wireless_table_structure_recognition_model_name="SLANeXt_wireless",
# 有线表格单元格检测
wired_table_cells_detection_model_name="RT-DETR-L_wired_table_cell_det",
# 无线表格单元格检测
wireless_table_cells_detection_model_name="RT-DETR-L_wireless_table_cell_det",
# wired_table_cells_detection_model_name="RT-DETR-L_wired_table_cell_det",
# wireless_table_structure_recognition_model_dir="G:\Python\YOLOv11\SLANeXt_wireless_infer"
# 是否加载并使用表格识别子产线
use_table_recognition=True,
# 是否加载印章识别子产线
use_seal_recognition=False,
# 是否加载并使用公式识别子产线
use_formula_recognition=False,
)
Beta Was this translation helpful? Give feedback.
All reactions