README.md 5.0 KB
Newer Older
W
WenmuZhou 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
# Table structure and content prediction

## 1. pipeline
The ocr of the table mainly contains three models
1. Single line text detection-DB
2. Single line text recognition-CRNN
3. Table structure and cell coordinate prediction-RARE

The table ocr flow chart is as follows

![tableocr_pipeline](../../doc/table/tableocr_pipeline.png)

1. The coordinates of single-line text is detected by DB model, and then sends it to the recognition model to get the recognition result.
2. The table structure and cell coordinates is predicted by RARE model.
3. The recognition result of the cell is combined by the coordinates, recognition result of the single line and the coordinates of the cell.
4. The cell recognition result and the table structure together construct the html string of the table.

## 2. How to use


### 2.1 Train
W
WenmuZhou 已提交
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

In this chapter, we only introduce the training of the table structure model, For model training of [text detection](../../doc/doc_en/detection_en.md) and [text recognition](../../doc/doc_en/recognition_en.md), please refer to the corresponding documents

#### data preparation  
The training data uses public data set [PubTabNet](https://arxiv.org/abs/1911.10683 ), Can be downloaded from the official [website](https://github.com/ibm-aur-nlp/PubTabNet) 。The PubTabNet data set contains about 500,000 images, as well as annotations in html format。

#### Start training  
*If you are installing the cpu version of paddle, please modify the `use_gpu` field in the configuration file to false*
```shell
# single GPU training
python3 tools/train.py -c configs/table/table_mv3.yml
# multi-GPU training
# Set the GPU ID used by the '--gpus' parameter.
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/table/table_mv3.yml
```

In the above instruction, use `-c` to select the training to use the `configs/table/table_mv3.yml` configuration file.
For a detailed explanation of the configuration file, please refer to [config](../../doc/doc_en/config_en.md).

#### load trained model and continue training

If you expect to load trained model and continue the training again, you can specify the parameter `Global.checkpoints` as the model path to be loaded.

```shell
python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./your/trained/model
```

**Note**: The priority of `Global.checkpoints` is higher than that of `Global.pretrain_weights`, that is, when two parameters are specified at the same time, the model specified by `Global.checkpoints` will be loaded first. If the model path specified by `Global.checkpoints` is wrong, the one specified by `Global.pretrain_weights` will be loaded.
W
WenmuZhou 已提交
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

### 2.2 Eval
First cd to the PaddleOCR/ppstructure directory

The table uses TEDS (Tree-Edit-Distance-based Similarity) as the evaluation metric of the model. Before the model evaluation, the three models in the pipeline need to be exported as inference models (we have provided them), and the gt for evaluation needs to be prepared. Examples of gt are as follows:
```json
{"PMC4289340_004_00.png": [["<html>", "<body>", "<table>", "<thead>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</thead>", "<tbody>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>",  "</tbody>", "</table>", "</body>", "</html>"], [[1, 4, 29, 13], [137, 4, 161, 13], [215, 4, 236, 13], [1, 17, 30, 27], [137, 17, 147, 27], [215, 17, 225, 27]], [["<b>", "F", "e", "a", "t", "u", "r", "e", "</b>"], ["<b>", "G", "b", "3", " ", "+", "</b>"], ["<b>", "G", "b", "3", " ", "-", "</b>"], ["<b>", "P", "a", "t", "i", "e", "n", "t", "s", "</b>"], ["6", "2"], ["4", "5"]]]}
```
In gt json, the key is the image name, the value is the corresponding gt, and gt is a list composed of four items, and each item is
1. HTML string list of table structure
2. The coordinates of each cell (not including the empty text in the cell)
3. The text information in each cell (not including the empty text in the cell)
4. The text information in each cell (including the empty text in the cell)

Use the following command to evaluate. After the evaluation is completed, the teds indicator will be output.
```python
W
WenmuZhou 已提交
66
python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --gt_path=path/to/gt.json
W
WenmuZhou 已提交
67 68 69 70 71 72 73
```


### 2.3 Inference
First cd to the PaddleOCR/ppstructure directory

```python
W
WenmuZhou 已提交
74
python3 table/predict_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table
W
WenmuZhou 已提交
75
```
W
opt doc  
WenmuZhou 已提交
76
After running, the excel sheet of each picture will be saved in the directory specified by the output field