Skip to content

P
PaddleOCR

s920243400 / PaddleOCR
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleOCR

通知 1
Star 1
Fork 0
P
PaddleOCR
提交 73ca6c2e 编写于 作者: 文幕地方's avatar 文幕地方
浏览文件
操作

add PP-Structurev2 to hubserving

上级 4ac17fca
显示空白变更内容
内联 并排
Showing with 63 addition and 73 deletion
deploy/hubserving/readme.md
浏览文件 @ 73ca6c2e
......@@ -59,6 +59,7 @@ pip3 install paddlehub==2.1.0 --upgrade -i https://mirror.baidu.com/pypi/simple
检测模型：./inference/ch_PP-OCRv3_det_infer/
识别模型：./inference/ch_PP-OCRv3_rec_infer/
方向分类器：./inference/ch_ppocr_mobile_v2.0_cls_infer/
版面分析模型：
表格结构识别模型：./inference/en_ppocr_mobile_v2.0_table_structure_infer/
```
......@@ -172,7 +173,7 @@ hub serving start -c deploy/hubserving/ocr_system/config.json
## 3. 发送预测请求
配置好服务端，可使用以下命令发送预测请求，获取预测结果:
```python tools/test_hubserving.py server_url image_path```
```python tools/test_hubserving.py --server_url=server_url --image_dir=image_path```
需要给脚本传递2个参数：
- **server_url**：服务地址，格式为
......
deploy/hubserving/readme_en.md
浏览文件 @ 73ca6c2e
......@@ -61,6 +61,7 @@ Before installing the service module, you need to prepare the inference model an
text detection model: ./inference/ch_PP-OCRv3_det_infer/
text recognition model: ./inference/ch_PP-OCRv3_rec_infer/
text angle classifier: ./inference/ch_ppocr_mobile_v2.0_cls_infer/
layout parse model:
tanle recognition: ./inference/en_ppocr_mobile_v2.0_table_structure_infer/
```
......@@ -177,7 +178,7 @@ hub serving start -c deploy/hubserving/ocr_system/config.json
## 3. Send prediction requests
After the service starts, you can use the following command to send a prediction request to obtain the prediction result:
```shell
python tools/test_hubserving.py server_url image_path
python tools/test_hubserving.py --server_url=server_url --image_dir=image_path
```
Two parameters need to be passed to the script:
......
deploy/hubserving/structure_system/module.py
浏览文件 @ 73ca6c2e
......@@ -119,7 +119,7 @@ class StructureSystem(hub.Module):
all_results.append([])
continue
starttime = time.time()
res = self.table_sys(img)
res, _ = self.table_sys(img)
elapse = time.time() - starttime
logger.info("Predict time: {}".format(elapse))
......@@ -144,6 +144,6 @@ class StructureSystem(hub.Module):
if __name__ == '__main__':
structure_system = StructureSystem()
structure_system._initialize()
image_path = ['./doc/table/1.png']
image_path = ['./ppstructure/docs/table/1.png']
res = structure_system.predict(paths=image_path)
print(res)
deploy/hubserving/structure_system/params.py
浏览文件 @ 73ca6c2e
......@@ -23,8 +23,10 @@ def read_params():
cfg = table_read_params()
# params for layout parser model
cfg.layout_path_model = 'lp://PubLayNet/ppyolov2_r50vd_dcn_365e_publaynet/config'
cfg.layout_label_map = None
cfg.layout_model_dir = ''
cfg.layout_dict_path = './ppocr/utils/dict/layout_publaynet_dict.txt'
cfg.layout_score_threshold = 0.5
cfg.layout_nms_threshold = 0.5
cfg.mode = 'structure'
cfg.output = './output'
......
deploy/hubserving/structure_table/module.py
浏览文件 @ 73ca6c2e
......@@ -118,11 +118,11 @@ class TableSystem(hub.Module):
all_results.append([])
continue
starttime = time.time()
pred_html = self.table_sys(img)
res, _ = self.table_sys(img)
elapse = time.time() - starttime
logger.info("Predict time: {}".format(elapse))
all_results.append({'html': pred_html})
all_results.append({'html': res['html']})
return all_results
@serving
......@@ -138,6 +138,6 @@ class TableSystem(hub.Module):
if __name__ == '__main__':
table_system = TableSystem()
table_system._initialize()
image_path = ['./doc/table/table.jpg']
image_path = ['./ppstructure/docs/table/table.jpg']
res = table_system.predict(paths=image_path)
print(res)
doc/doc_en/quickstart_en.md
浏览文件 @ 73ca6c2e
......@@ -51,12 +51,6 @@ pip install "paddleocr>=2.0.1" # Recommend to use version 2.0.1+
Reference: [Solve shapely installation on windows](https://stackoverflow.com/questions/44398265/install-shapely-oserror-winerror-126-the-specified-module-could-not-be-found)
- **For layout analysis users**, run the following command to install **Layout-Parser**
```bash
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
```
<a name="2-easy-to-use"></a>
## 2. Easy-to-Use
......
ppstructure/docs/installation.md
浏览文件 @ 73ca6c2e
- [快速安装](#快速安装)
- [1. PaddlePaddle 和 PaddleOCR](#1-paddlepaddle-和-paddleocr)
- [2. 安装其他依赖](#2-安装其他依赖)
- [2.1 版面分析所需 Layout-Parser](#21-版面分析所需--layout-parser)
- [2.2 VQA所需依赖](#22--vqa所需依赖)
- [2.1 VQA所需依赖](#21--vqa所需依赖)
# 快速安装
......@@ -12,14 +11,7 @@
## 2. 安装其他依赖
### 2.1 版面分析所需 Layout-Parser
Layout-Parser 可通过如下命令安装
```bash
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
```
### 2.2 VQA所需依赖
### 2.1 VQA所需依赖
* paddleocr
```bash
......
ppstructure/docs/quickstart.md
浏览文件 @ 73ca6c2e
# PP-Structure 快速开始
- [1. 安装依赖包](#1)
- [2. 便捷使用](#2)
- [2.1 命令行使用](#21)
- [2.1.1 版面分析+表格识别](#211)
- [2.1.2 版面分析](#212)
- [2.1.3 表格识别](#213)
- [2.1.4 DocVQA](#214)
- [2.2 代码使用](#22)
- [2.2.1 版面分析+表格识别](#221)
- [2.2.2 版面分析](#222)
- [2.2.3 表格识别](#223)
- [2.2.4 DocVQA](#224)
- [2.3 返回结果说明](#23)
- [2.3.1 版面分析+表格识别](#231)
- [2.3.2 DocVQA](#232)
- [2.4 参数说明](#24)
- [1. 安装依赖包](#1-安装依赖包)
- [2. 便捷使用](#2-便捷使用)
- [2.1 命令行使用](#21-命令行使用)
- [2.1.1 版面分析+表格识别](#211-版面分析表格识别)
- [2.1.2 版面分析](#212-版面分析)
- [2.1.3 表格识别](#213-表格识别)
- [2.1.4 DocVQA](#214-docvqa)
- [2.2 代码使用](#22-代码使用)
- [2.2.1 版面分析+表格识别](#221-版面分析表格识别)
- [2.2.2 版面分析](#222-版面分析)
- [2.2.3 表格识别](#223-表格识别)
- [2.2.4 DocVQA](#224-docvqa)
- [2.3 返回结果说明](#23-返回结果说明)
- [2.3.1 版面分析+表格识别](#231-版面分析表格识别)
- [2.3.2 DocVQA](#232-docvqa)
- [2.4 参数说明](#24-参数说明)
<a name="1"></a>
......@@ -24,8 +24,6 @@
```bash
# 安装 paddleocr，推荐使用2.5+版本
pip3 install "paddleocr>=2.5"
# 安装 版面分析依赖包layoutparser（如不需要版面分析功能，可跳过）
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
# 安装 DocVQA依赖包paddlenlp（如不需要DocVQA功能，可跳过）
pip install paddlenlp
......
ppstructure/docs/quickstart_en.md
浏览文件 @ 73ca6c2e
# PP-Structure Quick Start
- [1. Install package](#1)
- [2. Use](#2)
- [2.1 Use by command line](#21)
- [2.1.1 layout analysis + table recognition](#211)
- [2.1.2 layout analysis](#212)
- [2.1.3 table recognition](#213)
- [2.1.4 DocVQA](#214)
- [2.2 Use by code](#22)
- [2.2.1 layout analysis + table recognition](#221)
- [2.2.2 layout analysis](#222)
- [2.2.3 table recognition](#223)
- [2.2.4 DocVQA](#224)
- [2.3 Result description](#23)
- [2.3.1 layout analysis + table recognition](#231)
- [2.3.2 DocVQA](#232)
- [2.4 Parameter Description](#24)
- [1. Install package](#1-install-package)
- [2. Use](#2-use)
- [2.1 Use by command line](#21-use-by-command-line)
- [2.1.1 layout analysis + table recognition](#211-layout-analysis--table-recognition)
- [2.1.2 layout analysis](#212-layout-analysis)
- [2.1.3 table recognition](#213-table-recognition)
- [2.1.4 DocVQA](#214-docvqa)
- [2.2 Use by code](#22-use-by-code)
- [2.2.1 layout analysis + table recognition](#221-layout-analysis--table-recognition)
- [2.2.2 layout analysis](#222-layout-analysis)
- [2.2.3 table recognition](#223-table-recognition)
- [2.2.4 DocVQA](#224-docvqa)
- [2.3 Result description](#23-result-description)
- [2.3.1 layout analysis + table recognition](#231-layout-analysis--table-recognition)
- [2.3.2 DocVQA](#232-docvqa)
- [2.4 Parameter Description](#24-parameter-description)
<a name="1"></a>
......@@ -24,8 +24,6 @@
```bash
# Install paddleocr, version 2.5+ is recommended
pip3 install "paddleocr>=2.5"
# Install layoutparser (if you do not use the layout analysis, you can skip it)
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
# Install the DocVQA dependency package paddlenlp (if you do not use the DocVQA, you can skip it)
pip install paddlenlp
......
ppstructure/predict_system.py
浏览文件 @ 73ca6c2e
......@@ -43,6 +43,7 @@ logger = get_logger()
class StructureSystem(object):
def __init__(self, args):
self.mode = args.mode
self.recovery = args.recovery
if self.mode == 'structure':
if not args.show_log:
logger.setLevel(logging.INFO)
......@@ -110,7 +111,7 @@ class StructureSystem(object):
time_dict['rec'] += table_time_dict['rec']
else:
if self.text_system is not None:
if args.recovery:
if self.recovery:
wht_im = np.ones(ori_im.shape, dtype=ori_im.dtype)
wht_im[y1:y2, x1:x2, :] = roi_img
filter_boxes, filter_rec_res, ocr_time_dict = self.text_system(
......@@ -133,7 +134,7 @@ class StructureSystem(object):
for token in style_token:
if token in rec_str:
rec_str = rec_str.replace(token, '')
if not args.recovery:
if not self.recovery:
box += [x1, y1]
res.append({
'text': rec_str,
......
ppstructure/table/predict_table.py
浏览文件 @ 73ca6c2e
......@@ -101,7 +101,7 @@ class TableSystem(object):
start = time.time()
structure_res, elapse = self._structure(copy.deepcopy(img))
result['cell_bbox'] = structure_res[1]
result['cell_bbox'] = structure_res[1].tolist()
time_dict['table'] = elapse
dt_boxes, rec_res, det_elapse, rec_elapse = self._ocr(
......
ppstructure/utility.py
浏览文件 @ 73ca6c2e
......@@ -38,14 +38,17 @@ def init_args():
parser.add_argument(
"--layout_dict_path",
type=str,
default="../ppocr/utils/dict/layout_pubalynet_dict.txt")
default="../ppocr/utils/dict/layout_publaynet_dict.txt")
parser.add_argument(
"--layout_score_threshold",
type=float,
default=0.5,
help="Threshold of score.")
parser.add_argument(
"--layout_nms_threshold", type=float, default=0.5, help="Threshold of nms.")
"--layout_nms_threshold",
type=float,
default=0.5,
help="Threshold of nms.")
# params for vqa
parser.add_argument("--vqa_algorithm", type=str, default='LayoutXLM')
parser.add_argument("--ser_model_dir", type=str)
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或