forgetest5
/
PaddleOCR
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

fix errors and add pretrain_model

This commit is contained in:
Jethong 2021-04-07 12:26:05 +08:00
parent f698142542
commit 6824db26e7
8 changed files with 23 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
doc/doc_ch/e2e.md
View File

@ -31,7 +31,7 @@
|- rgb/ total_text数据集的训练数据
|- gt_0.png
| ...
|-poly/ total_text数据集的测试标注
|- poly/ total_text数据集的测试标注
|- gt_0.txt
| ...
```
@ -52,19 +52,11 @@
您可以根据需求使用[PaddleClas](https://github.com/PaddlePaddle/PaddleClas/tree/master/ppcls/modeling/architectures)中的模型更换backbone。
```shell
cd PaddleOCR/
下载ResNet50_vd的预训练模型
wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_ssld_pretrained.tar
下载ResNet50_vd的动态图预训练模型
wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams
# 解压预训练模型文件，以ResNet50_vd为例
tar -xf ./pretrain_models/ResNet50_vd_ssld_pretrained.tar ./pretrain_models/
# 注：正确解压backbone预训练权重文件后，文件夹下包含众多以网络层命名的权重文件，格式如下：
./pretrain_models/ResNet50_vd_ssld_pretrained/
└─ conv_last_bn_mean
└─ conv_last_bn_offset
└─ conv_last_bn_scale
└─ conv_last_bn_variance
└─ ......
./pretrain_models/
└─ ResNet50_vd_ssld_pretrained.pdparams
```
@ -74,11 +66,9 @@ tar -xf ./pretrain_models/ResNet50_vd_ssld_pretrained.tar ./pretrain_models/
```shell
# 单机单卡训练 e2e 模型
python3 tools/train.py -c configs/e2e/e2e_r50_vd_pg.yml \
-o Global.pretrain_weights=./pretrain_models/ResNet50_vd_ssld_pretrained/ Global.load_static_weights=True
python3 tools/train.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./pretrain_models/ResNet50_vd_ssld_pretrained Global.load_static_weights=False
# 单机多卡训练，通过 --gpus 参数设置使用的GPU ID
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/e2e/e2e_r50_vd_pg.yml \
-o Global.pretrain_weights=./pretrain_models/ResNet50_vd_ssld_pretrained/ Global.load_static_weights=True
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./pretrain_models/ResNet50_vd_ssld_pretrained Global.load_static_weights=False
```

13
doc/doc_ch/inference.md
View File

@ -369,9 +369,9 @@ Predicts of ./doc/imgs_words/korean/1.jpg:('바탕으로', 0.9948904)
<a name="PGNet端到端模型推理"></a>
### 1. PGNet端到端模型推理
#### (1). 四边形文本检测模型（ICDAR2015）
首先将PGNet端到端训练过程中保存的模型，转换成inference model。以基于Resnet50_vd骨干网络，在ICDAR2015英文数据集训练的模型为例([模型下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar))，可以使用如下命令进行转换：
首先将PGNet端到端训练过程中保存的模型，转换成inference model。以基于Resnet50_vd骨干网络，在ICDAR2015英文数据集训练的模型为例([模型下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/en_server_pgnetA.tar))，可以使用如下命令进行转换：
```
python3 tools/export_model.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./det_r50_vd_sast_icdar15_v2.0_train/best_accuracy Global.load_static_weights=False Global.save_inference_dir=./inference/e2e
python3 tools/export_model.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./en_server_pgnetA/iter_epoch_450 Global.load_static_weights=False Global.save_inference_dir=./inference/e2e
```
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"`**，可以执行如下命令：
```
@ -382,15 +382,10 @@ python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/im
![](../imgs_results/e2e_res_img_10_pgnet.jpg)
#### (2). 弯曲文本检测模型（Total-Text）
首先将PGNet端到端训练过程中保存的模型，转换成inference model。以基于Resnet50_vd骨干网络，在Total-Text英文数据集训练的模型为例（[模型下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_totaltext_v2.0_train.tar))，可以使用如下命令进行转换：
```
python3 tools/export_model.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./det_r50_vd_sast_totaltext_v2.0_train/best_accuracy Global.load_static_weights=False Global.save_inference_dir=./inference/e2e
```
和四边形文本检测模型共用一个推理模型
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"`，同时，还需要增加参数`--e2e_pgnet_polygon=True`，**可以执行如下命令：
```
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=True
python3.7 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=True
```
可视化文本端到端结果默认保存到`./inference_results`文件夹里面，结果文件的名称前缀为'e2e_res'。结果示例如下：

4
ppocr/data/imaug/pg_process.py
View File

@ -27,7 +27,7 @@ class PGProcessTrain(object):
tcl_len,
batch_size=14,
min_crop_size=24,
min_text_size=10,
min_text_size=4,
max_text_size=512,
**kwargs):
self.tcl_len = tcl_len
@ -197,7 +197,6 @@ class PGProcessTrain(object):
for selected_poly in selected_polys:
txts_tmp.append(txts[selected_poly])
txts = txts_tmp
# print(1111)
return im[ymin: ymax + 1, xmin: xmax + 1, :], \
polys[selected_polys], tags[selected_polys], hv_tags[selected_polys], txts
else:
@ -309,7 +308,6 @@ class PGProcessTrain(object):
cv2.fillPoly(direction_map,
quad.round().astype(np.int32)[np.newaxis, :, :],
direction_label)
cv2.imwrite("output/{}.png".format(k), direction_map * 255.0)
k += 1
return direction_map

9
ppocr/data/pgnet_dataset.py
View File

@ -67,10 +67,7 @@ class PGDataSet(Dataset):
np.array(
list(poly), dtype=np.float32).reshape(-1, 2))
txts.append(txt)
if txt == '###':
txt_tags.append(True)
else:
txt_tags.append(False)
txt_tags.append(txt == '###')
return np.array(list(map(np.array, text_polys))), \
np.array(txt_tags, dtype=np.bool), txts
@ -84,8 +81,8 @@ class PGDataSet(Dataset):
for ext in [
'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'JPG'
]:
if os.path.exists(os.path.join(img_dir, info_list[0] + ext)):
img_path = os.path.join(img_dir, info_list[0] + ext)
if os.path.exists(os.path.join(img_dir, info_list[0] + "." + ext)):
img_path = os.path.join(img_dir, info_list[0] + "." + ext)
break
if img_path == '':

2
ppocr/losses/e2e_pg_loss.py
View File

@ -20,7 +20,7 @@ from paddle import nn
import paddle
from .det_basic_loss import DiceLoss
from ppocr.utils.e2e_utils.extract_batchsize import *
from ppocr.utils.e2e_utils.extract_batchsize import pre_process
class PGLoss(nn.Layer):

4
ppocr/metrics/e2e_metric.py
View File

@ -18,8 +18,8 @@ from __future__ import print_function
__all__ = ['E2EMetric']
from ppocr.utils.e2e_metric.Deteval import *
from ppocr.utils.e2e_utils.extract_textpoint import *
from ppocr.utils.e2e_metric.Deteval import get_socre, combine_results
from ppocr.utils.e2e_utils.extract_textpoint import get_dict
class E2EMetric(object):

3
requirements.txt
View File

@ -7,4 +7,5 @@ opencv-python==4.2.0.32
tqdm
numpy
visualdl
python-Levenshtein
python-Levenshtein
opencv-contrib-python

6
tools/infer/predict_e2e.py
View File

@ -34,7 +34,7 @@ from ppocr.postprocess import build_post_process
logger = get_logger()
class TextE2e(object):
class TextE2E(object):
def __init__(self, args):
self.args = args
self.e2e_algorithm = args.e2e_algorithm
@ -130,7 +130,7 @@ class TextE2e(object):
if __name__ == "__main__":
args = utility.parse_args()
image_file_list = get_image_file_list(args.image_dir)
text_detector = TextE2e(args)
text_detector = TextE2E(args)
count = 0
total_time = 0
draw_img_save = "./inference_results"
@ -151,7 +151,7 @@ if __name__ == "__main__":
src_im = utility.draw_e2e_res(points, strs, image_file)
img_name_pure = os.path.split(image_file)[-1]
img_path = os.path.join(draw_img_save,
"e2e_res_{}".format(img_name_pure))
"e2e_res_{}_pgnet".format(img_name_pure))
cv2.imwrite(img_path, src_im)
logger.info("The visualized image saved in {}".format(img_path))
if count > 1: