PaddleOCR Freezes on M2 MacOS Ventura version 13.5

[pranay-009]

I installed

• PyMuPDF-1.20.2 and paddleocr-2.7.0 and PaddlePaddle 2.3.2
• I was using python 3.8.18
• System Information :Mac M2 MacOS Ventura version 13.5

I had to force and stop the process

code sample

from paddleocr import PaddleOCR
ocr = PaddleOCR(use_angle_cls=True, lang='en') # need to run only once to download and load model into memory
img_path = '/Users/pranaysaha/Developer/OCR/sample.jpeg'
result = ocr.ocr(img_path, cls=True) 

[pavan5849]

I'm also facing same issue
My configuration:
Python 3.9.6
Macos: Sonoma 14.1.1
Apple M2 pro
imageio==2.10.1
opencv-contrib-python==4.6.0.66
opencv-python-headless==4.6.0.66
protobuf==3.20.0
paddleocr==2.7.0.3
paddlepaddle==2.4.0

[nvinayvarma189]

I'm also facing same issue. My configuration:

Python 3.10.12
Macos: Sonoma 14.1.2
Apple M2 pro

opencv-contrib-python==4.6.0.66
opencv-python==4.6.0.66
protobuf==4.25.1
paddleocr==2.7.0.3
paddlepaddle==2.5.1
pymupdf==1.20.0

I tried the same code as @pranay-009 and the execution freezes after this:

[2023/12/05 15:17:08] ppocr DEBUG: Namespace(help='==SUPPRESS==', use_gpu=False, use_xpu=False, use_npu=False, ir_optim=True, use_tensorrt=False, min_subgraph_size=15, precision='fp32', gpu_mem=500, gpu_id=0, image_dir=None, page_num=0, det_algorithm='DB', det_model_dir='/Users/roronoa/.paddleocr/whl/det/en/en_PP-OCRv3_det_infer', det_limit_side_len=960, det_limit_type='max', det_box_type='quad', det_db_thresh=0.3, det_db_box_thresh=0.6, det_db_unclip_ratio=1.5, max_batch_size=10, use_dilation=False, det_db_score_mode='fast', det_east_score_thresh=0.8, det_east_cover_thresh=0.1, det_east_nms_thresh=0.2, det_sast_score_thresh=0.5, det_sast_nms_thresh=0.2, det_pse_thresh=0, det_pse_box_thresh=0.85, det_pse_min_area=16, det_pse_scale=1, scales=[8, 16, 32], alpha=1.0, beta=1.0, fourier_degree=5, rec_algorithm='SVTR_LCNet', rec_model_dir='/Users/roronoa/.paddleocr/whl/rec/en/en_PP-OCRv4_rec_infer', rec_image_inverse=True, rec_image_shape='3, 48, 320', rec_batch_num=6, max_text_length=25, rec_char_dict_path='/Users/roronoa/.local/share/virtualenvs/hemkund-backend-w6cFaI3p/lib/python3.10/site-packages/paddleocr/ppocr/utils/en_dict.txt', use_space_char=True, vis_font_path='./doc/fonts/simfang.ttf', drop_score=0.5, e2e_algorithm='PGNet', e2e_model_dir=None, e2e_limit_side_len=768, e2e_limit_type='max', e2e_pgnet_score_thresh=0.5, e2e_char_dict_path='./ppocr/utils/ic15_dict.txt', e2e_pgnet_valid_set='totaltext', e2e_pgnet_mode='fast', use_angle_cls=True, cls_model_dir='/Users/roronoa/.paddleocr/whl/cls/ch_ppocr_mobile_v2.0_cls_infer', cls_image_shape='3, 48, 192', label_list=['0', '180'], cls_batch_num=6, cls_thresh=0.9, enable_mkldnn=False, cpu_threads=10, use_pdserving=False, warmup=False, sr_model_dir=None, sr_image_shape='3, 32, 128', sr_batch_num=1, draw_img_save_dir='./inference_results', save_crop_res=False, crop_res_save_dir='./output', use_mp=False, total_process_num=1, process_id=0, benchmark=False, save_log_path='./log_output/', show_log=True, use_onnx=False, output='./output', table_max_len=488, table_algorithm='TableAttn', table_model_dir=None, merge_no_span_structure=True, table_char_dict_path=None, layout_model_dir=None, layout_dict_path=None, layout_score_threshold=0.5, layout_nms_threshold=0.5, kie_algorithm='LayoutXLM', ser_model_dir=None, re_model_dir=None, use_visual_backbone=True, ser_dict_path='../train_data/XFUND/class_list_xfun.txt', ocr_order_method=None, mode='structure', image_orientation=False, layout=True, table=True, ocr=True, recovery=False, use_pdf2docx_api=False, invert=False, binarize=False, alphacolor=(255, 255, 255), lang='en', det=True, rec=True, type='ocr', ocr_version='PP-OCRv4', structure_version='PP-StructureV2')

[sanduluca]

I have same issue. Does anyone have a solution for this ?

Apple M2 Pro, 16 GB, Sonoma 14.0
Python 3.11.6

numpy==1.26.2
opencv-contrib-python==4.8.1.78
opencv-python==4.8.1.78
paddleocr==2.6.0.1
paddlepaddle==2.5.2
pandas==2.1.4
Pillow==10.1.0
protobuf==4.23.4
PyMuPDF==1.21.1

[PterX]

I have same issue. Does anyone have a solution for this ?

[linjson]

I have same issue,

[BB86818BB]

I have same issue,
m1 pro
paddlepaddle 2.6.0
paddleocr 2.6.0.1
Python 3.8.18
opencv-contrib-python 4.6.0.66/4.9.0.80
opencv-python 4.6.0.66/4.9.0.80

[linjson]

先用 rapidocr平替吧,用 onnx推理,模型还是用paddle的,在m2上可用

[vikasr111]

I am facing the same issue on Macbook M3 Pro.
paddlepaddle==2.6.0
paddleocr==2.7.0.3

Macos: Sonoma 14.1.2
Python 3.10.13 (But it doesn't matter, tested on multiple Python version).
I hope there's some solution to running this on Mac with Apple Silicon.

[benx13]

same issue running on M1 Pro

[jzhang533]

cc: @michaelowenliu @dyning @tink2123
This should be a high-priority issue that needs the maintainers' attention.

[jzhang533]

I can run the following script on my Apple M1, macOS Monterey, python 3.9.13 without freezing issue.

from paddleocr import PaddleOCR
ocr = PaddleOCR(use_angle_cls=True, lang='en')
img_path = './1.jpeg'
result = ocr.ocr(img_path, cls=True)

I have the following packages installed:

paddleocr 2.7.0.3
paddlepaddle 2.6.0

I'm unclear about the differences between the various M-series chips, but I'm hoping they'll be compatible.
Would anyone be willing to upgrade PaddlePaddle to version 2.6.0 and share the result?

[benx13]

I can run the following script on my Apple M1, macOS Monterey, python 3.9.13 without freezing issue.

from paddleocr import PaddleOCR
ocr = PaddleOCR(use_angle_cls=True, lang='en')
img_path = './1.jpeg'
result = ocr.ocr(img_path, cls=True)

I have the following packages installed:

paddleocr 2.7.0.3 paddlepaddle 2.6.0

I'm unclear about the differences between the various M-series chips, but I'm hoping they'll be compatible. Would anyone be willing to upgrade PaddlePaddle to version 2.6.0 and share the result?

I've got
python 3.9.6
Sonoma 14.2.1

paddleocr 2.7.0.3
paddlepaddle 2.6.0
PyMuPDF 1.20.2

still freezing

[sunzhongkai588]

I've got python 3.9.6 Sonoma 14.2.1

paddleocr 2.7.0.3 paddlepaddle 2.6.0 PyMuPDF 1.20.2

still freezing

I'm using Python 3.9.13 and Sonoma 14.0, with all other package versions matching yours.

I can run the program successfully. It's possible there may be an issue with your image. Could you share an example image？

[benx13]

I've got python 3.9.6 Sonoma 14.2.1
paddleocr 2.7.0.3 paddlepaddle 2.6.0 PyMuPDF 1.20.2
still freezing

I'm using Python 3.9.13 and Sonoma 14.0, with all other package versions matching yours.

I can run the program successfully. It's possible there may be an issue with your image. Could you share an example image？

you can use:
img_path = './imgs_en/img_12.jpg'
from the official PaddleOCR repository

[jzhang533]

you can use:
img_path = './imgs_en/img_12.jpg'
from the official PaddleOCR repository

still can't reproduce this freezing issue, can you share result of running this :
python -mplatform 'platform.platform()'

[R4Ajeti]

@jzhang533 I have same issue as @benx13
apple M3 pro
python version: 3.12.1
paddleocr==2.6.0.1
paddlepaddle==2.6.0
PyMuPDF==1.23.26
opencv-contrib-python==4.6.0.66
opencv-python==4.6.0.66

python -mplatform 'platform.platform()'
gives this one:
macOS-14.4-arm64-arm-64bit

[kilimchoi]

@jzhang533 experiencing the same thing.
I have:

Apple M1 Ventura 13.1
Python 3.10.12
paddleocr==2.7.0.3
paddlepaddle==2.6.0
pymupdf==1.20.2

python -mplatform 'platform.platform()' returns: macOS-13.1-arm64-arm-64bit

[benx13]

you can use:
img_path = './imgs_en/img_12.jpg'
from the official PaddleOCR repository

still can't reproduce this freezing issue, can you share result of running this : python -mplatform 'platform.platform()'

python -mplatform 'platform.platform()'

@jzhang533 it gives:
macOS-14.2.1-arm64-arm-64bit
references:
python 3.9.6
paddleocr 2.7.0.3
paddlepaddle 2.6.0
PyMuPDF 1.20.2
M1 Pro chip

[avakashsahoo-spry]

macOS-14.1.2-arm64-arm-64bit

Getting stuck in my mac M2 as well

[benx13]

@jzhang533 surprisingly, on a different Python version, it worked for me technically on the default version so the way I had it set up is I installed it on python 3.9 which is not my default Python my default is 3.12. When I installed it on python 3.12 it surprisingly works no issues.

[avakashsahoo-spry]

@jzhang533 surprisingly, on a different Python version, it worked for me technically on the default version so the way I had it set up is I installed it on python 3.9 which is not my default Python my default is 3.12. When I installed it on python 3.12 it surprisingly works no issues.

Gets stuck on 3.12 version as well

[R4Ajeti]

@jzhang533 surprisingly, on a different Python version, it worked for me technically on the default version so the way I had it set up is I installed it on python 3.9 which is not my default Python my default is 3.12. When I installed it on python 3.12 it surprisingly works no issues.

Gets stuck on 3.12 version as well

Ofc it does, I have debugged it fully and the conclusion: problem lies on paddlepaddle inference not on paddle ocr check this thread for more info:
#11706

[vikasr111]

For all who are still facing this issue, I have managed to run to successfully on Mac M3 Pro using Docker. Here's the Dockerfile you can use to build and run PaddleOCR. Hope this helps.

# Base Stage: Install common dependencies
FROM python:3.10-slim-bullseye

# Set the working directory in the container to /app
WORKDIR /app

# Set PYTHONPATH to include the /app directory
ENV PYTHONPATH="${PYTHONPATH}:/app"

# Add the rest of the current directory contents into the container at /app
COPY . .

# Install dependencies for OCR
RUN apt-get update \
    && apt-get install --no-install-recommends -y ffmpeg libsm6 libxext6 build-essential \
    # Install your Python packages that require compilation here
    && pip install --no-cache-dir python-dotenv==1.0.1 \
    && pip install --no-cache-dir paddlepaddle==2.6.0 \
    && pip install --no-cache-dir paddleocr==2.7.0.3 \
    # Cleanup: Remove build-essential and cleanup apt cache in the same command
    && apt-get purge -y build-essential \
    && apt-get autoremove -y \
    && rm -rf /var/lib/apt/lists/* /root/.cache/pip

To build and run this Docker image:

docker build -t paddle-ocr
docker exec -it paddle-ocr /bin/bash

[jinbinhan]

同样在M2上卡住了

[fuji246]

For all who are still facing this issue, I have managed to run to successfully on Mac M3 Pro using Docker. Here's the Dockerfile you can use to build and run PaddleOCR. Hope this helps.

# Base Stage: Install common dependencies
FROM python:3.10-slim-bullseye

# Set the working directory in the container to /app
WORKDIR /app

# Set PYTHONPATH to include the /app directory
ENV PYTHONPATH="${PYTHONPATH}:/app"

# Add the rest of the current directory contents into the container at /app
COPY . .

# Install dependencies for OCR
RUN apt-get update \
    && apt-get install --no-install-recommends -y ffmpeg libsm6 libxext6 build-essential \
    # Install your Python packages that require compilation here
    && pip install --no-cache-dir python-dotenv==1.0.1 \
    && pip install --no-cache-dir paddlepaddle==2.6.0 \
    && pip install --no-cache-dir paddleocr==2.7.0.3 \
    # Cleanup: Remove build-essential and cleanup apt cache in the same command
    && apt-get purge -y build-essential \
    && apt-get autoremove -y \
    && rm -rf /var/lib/apt/lists/* /root/.cache/pip

To build and run this Docker image:

docker build -t paddle-ocr
docker exec -it paddle-ocr /bin/bash

not working

root@e310947389a8:/app# paddleocr --image_dir /mnt/UML\ diagram.jpg --lang en --use_gpu false
download https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar to /root/.paddleocr/whl/det/en/en_PP-OCRv3_det_infer/en_PP-OCRv3_det_infer.tar
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4.00M/4.00M [00:09<00:00, 422kiB/s]


--------------------------------------
C++ Traceback (most recent call last):
--------------------------------------
0   inflateReset2

----------------------
Error Message Summary:
----------------------
FatalError: `Segmentation fault` is detected by the operating system.
  [TimeInfo: *** Aborted at 1712249711 (unix time) try "date -d @1712249711" if you are using GNU date ***]
  [SignalInfo: *** SIGSEGV (@0x18) received by PID 39 (TID 0xffffa023ed40) from PID 24 ***]

Segmentation fault

[SWHL]

如果问题依然存在，可以参考link

If the problem still exists, you could try to solve it according this link

[nickchen1998]

If you're encountering this issue on an Apple M-series chip (based on other comments, it seems other operating systems don’t have this problem), and it's happening specifically within a Celery worker, you can use the following command to run it:

celery -A celery_app.celery worker --loglevel=info --pool solo

The key is to include --pool solo, which forces Celery to run in single-threaded mode. This approach allows it to function properly during local development.