From d77db841b043f09a6e5b14ee0fcea4d636778bf1 Mon Sep 17 00:00:00 2001 From: littletomatodonkey Date: Thu, 26 Aug 2021 07:02:42 +0800 Subject: [PATCH] fix whl exit bug (#3806) * fix whl exit bug * fix recognition doc --- doc/doc_ch/recognition.md | 6 +++--- doc/doc_en/recognition_en.md | 6 +++--- ppocr/data/__init__.py | 12 +++++++----- 3 files changed, 13 insertions(+), 11 deletions(-) diff --git a/doc/doc_ch/recognition.md b/doc/doc_ch/recognition.md index 0ff0513a..3c79c03e 100644 --- a/doc/doc_ch/recognition.md +++ b/doc/doc_ch/recognition.md @@ -185,11 +185,11 @@ python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs #### 2.1 数据增强 -PaddleOCR提供了多种数据增强方式，如果您希望在训练时加入扰动，请在配置文件中设置 `distort: true`。 +PaddleOCR提供了多种数据增强方式，默认配置文件中已经添加了数据增广。 -默认的扰动方式有：颜色空间转换(cvtColor)、模糊(blur)、抖动(jitter)、噪声(Gasuss noise)、随机切割(random crop)、透视(perspective)、颜色反转(reverse)。 +默认的扰动方式有：颜色空间转换(cvtColor)、模糊(blur)、抖动(jitter)、噪声(Gasuss noise)、随机切割(random crop)、透视(perspective)、颜色反转(reverse)、TIA数据增广。 -训练过程中每种扰动方式以50%的概率被选择，具体代码实现请参考：[img_tools.py](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/ppocr/data/rec/img_tools.py) +训练过程中每种扰动方式以40%的概率被选择，具体代码实现请参考：[rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py) *由于OpenCV的兼容性问题，扰动操作暂时只支持Linux* diff --git a/doc/doc_en/recognition_en.md b/doc/doc_en/recognition_en.md index 634ec783..4a36db1e 100644 --- a/doc/doc_en/recognition_en.md +++ b/doc/doc_en/recognition_en.md @@ -177,11 +177,11 @@ python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs #### 2.1 Data Augmentation -PaddleOCR provides a variety of data augmentation methods. If you want to add disturbance during training, please set `distort: true` in the configuration file. +PaddleOCR provides a variety of data augmentation methods. All the augmentation methods are enabled by default. -The default perturbation methods are: cvtColor, blur, jitter, Gasuss noise, random crop, perspective, color reverse. +The default perturbation methods are: cvtColor, blur, jitter, Gasuss noise, random crop, perspective, color reverse, TIA augmentation. -Each disturbance method is selected with a 50% probability during the training process. For specific code implementation, please refer to: [img_tools.py](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/ppocr/data/rec/img_tools.py) +Each disturbance method is selected with a 40% probability during the training process. For specific code implementation, please refer to: [rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py) #### 2.2 Training diff --git a/ppocr/data/__init__.py b/ppocr/data/__init__.py index e860c5a6..0bb3d506 100644 --- a/ppocr/data/__init__.py +++ b/ppocr/data/__init__.py @@ -49,14 +49,12 @@ def term_mp(sig_num, frame): os.killpg(pgid, signal.SIGKILL) -signal.signal(signal.SIGINT, term_mp) -signal.signal(signal.SIGTERM, term_mp) - - def build_dataloader(config, mode, device, logger, seed=None): config = copy.deepcopy(config) - support_dict = ['SimpleDataSet', 'LMDBDataSet', 'PGDataSet', 'PubTabDataSet'] + support_dict = [ + 'SimpleDataSet', 'LMDBDataSet', 'PGDataSet', 'PubTabDataSet' + ] module_name = config[mode]['dataset']['name'] assert module_name in support_dict, Exception( 'DataSet only support {}'.format(support_dict)) @@ -96,4 +94,8 @@ def build_dataloader(config, mode, device, logger, seed=None): return_list=True, use_shared_memory=use_shared_memory) + # support exit using ctrl+c + signal.signal(signal.SIGINT, term_mp) + signal.signal(signal.SIGTERM, term_mp) + return data_loader -- GitLab