🔎 Search before asking

[x] I have searched the PaddleOCR Docs and found no similar bug report.
[x] I have searched the PaddleOCR Issues and found no similar bug report.
[x] I have searched the PaddleOCR Discussions and found no similar bug report.

🐛 Bug (问题描述)

我运行报错-------------------------------------- C++ Traceback (most recent call last):

0 phi::OneDNNContextThreadLocals::Body::~Body() 1 phi::OneDNNContext::Impl::ResetBlobMap(void*)

Error Message Summary:

FatalError: Process abort signal is detected by the operating system. [TimeInfo: *** Aborted at 1748315176 (unix time) try "date -d @1748315176" if you are using GNU date ***] [SignalInfo: *** SIGABRT (@0x3e90023bacb) received by PID 2341579 (TID 0x7f42dd89a280) from PID 2341579 ***]

🏃‍♂️ Environment (运行环境)

cuda11.8 python3.10 paddlepaddle-gpu==3.0.0 paddleocr=3.0.0

🌰 Minimal Reproducible Example (最小可复现问题的Demo)

from paddleocr import PaddleOCR import paddle import time import os import logging

设置日志记录器

logger = logging.getLogger() logger.setLevel(logging.INFO)

清空现有的处理器以避免重复

logger.handlers.clear()

添加新的处理器

ch = logging.StreamHandler() formatter = logging.Formatter("%(asctime)s - %(module)s - %(funcName)s - line:%(lineno)d - %(levelname)s - %(message)s") ch.setFormatter(formatter) logger.addHandler(ch)

pwd_path = os.path.abspath(os.path.dirname(file)) logger.info(f"PaddeleOCR is runing on path: {pwd_path}")

os.environ["CUDA_VISIBLE_DEVICES"] = "3"

use_cuda = paddle.is_compiled_with_cuda() # 检查paddle是否编译了 CUDA 支持 logger.info(f'use_cuda:{use_cuda}') if use_cuda: cuda_name = paddle.device.cuda.get_device_name() # 获取当前使用的 GPU 型号 logger.info(f'cuda_name:{cuda_name}')

t1 = time.time()

ocr = PaddleOCR(ocr_version="PP-OCRv5", use_doc_orientation_classify=False, # 文档方向分类模型默认true 检测整个文档图像的方向并纠正（即0度，90度，180度，270度）。 use_doc_unwarping=False, # 文本图像矫正模型默认true 纠正图像中的透视变形或弯曲，使文档看起来像是正面拍摄的平面图像。 use_textline_orientation=False, # 文本行方向分类模型默认true 识别文本行是水平排列还是垂直排列。 device="gpu" )

result = ocr.predict(01.png") t2 = time.time() print(f"ocr time {t2 - t1}")

for res in result: res.print() res.save_to_img("output") res.save_to_json("output")