Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

predict,最多只能识别8个图片,多了会暴显存 #937

Open
dengmingD opened this issue Sep 30, 2020 · 5 comments
Open

predict,最多只能识别8个图片,多了会暴显存 #937

dengmingD opened this issue Sep 30, 2020 · 5 comments
Assignees

Comments

@dengmingD
Copy link

hub:1.82
code:
images = [cv2.imread(image_path) for image_path in test_image]
#就是下面这行代码报错
result = ocr.recognize_text(images=images,use_gpu=True,
output_dir='ocr_result',
visualization=True
# box_thresh=0.5,
# text_thresh=0.5
)
error:
[2020-09-30 15:49:30,097] [ INFO] - Installing chinese_ocr_db_crnn_server module
[2020-09-30 15:49:30,176] [ INFO] - Module chinese_ocr_db_crnn_server already installed in C:\Users\1.paddlehub\modules\chinese_ocr_db_crnn_server
[2020-09-30 15:49:31,412] [ INFO] - Installing chinese_text_detection_db_server module-1.0.2
[2020-09-30 15:49:31,413] [ INFO] - Module chinese_text_detection_db_server-1.0.2 already installed in C:\Users\1.paddlehub\modules\chinese_text_detection_db_server
Traceback (most recent call last):
File "D:/project/new1/paddle/ocr_text.py", line 32, in
visualization=True
File "C:\Users\1.paddlehub\modules\chinese_ocr_db_crnn_server\module.py", line 271, in recognize_text
rec_results = self._recognize_text(img_crop_list)
File "C:\Users\1.paddlehub\modules\chinese_ocr_db_crnn_server\module.py", line 407, in _recognize_text
self.rec_predictor.zero_copy_run()
RuntimeError:


C++ Call Stacks (More useful to developers):

Windows not support stack backtrace yet.


Error Message Summary:

ResourceExhaustedError:

Out of memory error on GPU 0. Cannot allocate 183.047119MB memory on GPU 0, available memory is only 139.662499MB.

Please check whether there is any other process using GPU 0.

  1. If yes, please stop them, or start PaddlePaddle on another GPU.
  2. If no, please decrease the batch size of your model.

at (D:\1.8.5\paddle\paddle\fluid\memory\allocation\cuda_allocator.cc:69)

gpu:2070 super
cpu:10700k
显存没有一直被占用

@LDOUBLEV
Copy link

LDOUBLEV commented Sep 30, 2020

参考: #682
paddle 2.0 版本会修复

@FD-Liekkas
Copy link

参考: #682
paddle 2.0 版本会修复

您好,请问这个问题在paddle 2.0.0b0版本修复了吗?我在版本更新日志中没看到相关说明,还是说只有2.0正式版才会修复?麻烦您解答一下

@liweigu
Copy link

liweigu commented Jan 12, 2021

在 chinese_ocr_db_crnn_server 1.1.1 (paddlepaddle-gpu==2.0.0rc1.post110) 上测试还未解决。

@FD-Liekkas
Copy link

在 chinese_ocr_db_crnn_server 1.1.1 (paddlepaddle-gpu==2.0.0rc1.post110) 上测试还未解决。

我在python3.7.9+paddlepaddle-gpu==2.0.0rc1.post100-cp37上跑通了,没出现显存问题了

@SilenceBlue
Copy link

在 chinese_ocr_db_crnn_server 1.1.1 (paddlepaddle-gpu==2.0.0rc1.post110) 上测试还未解决。

我在python3.7.9+paddlepaddle-gpu==2.0.0rc1.post100-cp37上跑通了,没出现显存问题了

paddlepaddle==2.0.0rc1上测试还未解决

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants