------------------------------------
LayoutLM模型汇总
------------------------------------



下表汇总介绍了目前PaddleNLP支持的LayoutLM模型以及对应预训练权重。
关于模型的具体细节可以参考对应链接。

+----------------------------------------------------------------------------------+--------------+----------------------------------------------------------------------------------+
| Pretrained Weight                                                                | Language     | Details of the model                                                             |
+==================================================================================+==============+==================================================================================+
|``layoutlm-base-uncased``                                                         | English      | 12-layer, 768-hidden,                                                            |
|                                                                                  |              | 12-heads, 339M parameters.                                                       |
|                                                                                  |              | LayoutLm base uncased model.                                                     |
+----------------------------------------------------------------------------------+--------------+----------------------------------------------------------------------------------+
|``layoutlm-large-uncased``                                                        | English      | 24-layer, 1024-hidden,                                                           |
|                                                                                  |              | 16-heads, 51M parameters.                                                        |
|                                                                                  |              | LayoutLm large Uncased model.                                                    |
+----------------------------------------------------------------------------------+--------------+----------------------------------------------------------------------------------+