PaddleOCR/latest/quick_start #14035
Replies: 34 comments 64 replies
-
一个项目的首页为什么连个docker或者正常的安装命令都找不到了 |
Beta Was this translation helpful? Give feedback.
-
为啥训练完自己数据,转换为方便调用的形式,性能降了很多,是枝剪了很多吗? 比如最表面的,训练完后的ocr的rec,模型有180M+大小,转换后,只有十几M大小。(比如:python tools/export_model.py -c "./configs/rec/PP-OCRv3/ch_PP-OCRv3_rec.yml" -o Global.pretrained_model="./output/v3_en_mobile/iter_epoch_225.pdparams" Global.save_inference_dir="./inference_model/rec/") |
Beta Was this translation helpful? Give feedback.
-
模型转换的应该有问题了,转的哪个模型呀
…---- Replied Message ----
| From | ***@***.***> |
| Date | 11/03/2024 21:33 |
| To | PaddlePaddle/PaddleOCR ***@***.***> |
| Cc | SWHL ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [PaddlePaddle/PaddleOCR] PaddleOCR/latest/quick_start (Discussion #14035) |
为啥训练完自己数据,转换为方便调用的形式,性能降了很多,是枝剪了很多吗? 比如最表面的,训练完后的ocr的rec,模型有180M+大小,转换后,只有十几M大小。(比如:python tools/export_model.py -c "./configs/rec/PP-OCRv3/ch_PP-OCRv3_rec.yml" -o Global.pretrained_model="./output/v3_en_mobile/iter_epoch_225.pdparams" Global.save_inference_dir="./inference_model/rec/")
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
在Ubuntu上运行时,发现占用的内存一直没有释放,导致跑几张图片后内存就占满了是为什么呢 |
Beta Was this translation helpful? Give feedback.
-
[[[442.0, 173.0], [1169.0, 173.0], [1169.0, 225.0], [442.0, 225.0]], ['ACKNOWLEDGEMENTS', 0.99283075]] |
Beta Was this translation helpful? Give feedback.
-
求助,为什么安装了paddleocr以后一直在import(mac m1) |
Beta Was this translation helpful? Give feedback.
-
你好,请问这个支持离线构建部署吗? |
Beta Was this translation helpful? Give feedback.
-
CPU版本推理需要4秒以上吗? |
Beta Was this translation helpful? Give feedback.
-
如果提供这些图片文件就更好啦,比如“'PaddleOCR/doc/imgs_en/img_12.jpg'”还有其他在代码中出现的 |
Beta Was this translation helpful? Give feedback.
-
请问有开源的 可以识别中文字符的模型吗 |
Beta Was this translation helpful? Give feedback.
-
为什么检测+识别+分类,输出里没有方向分类信息? |
Beta Was this translation helpful? Give feedback.
-
修改一下吧: |
Beta Was this translation helpful? Give feedback.
-
打开PP-OCRv4 在线体验地址,使用通用OCR(高效率和高精度模式都试过)测试识别合同封面和内页图片,会把文字水印、印章中的文字识别出来(这些文字其实是不需要的),输出的识别结果是正常文字和不需要的文字交叉混杂在一起,请问能增加剔除不需要的文字的处理吗(还是说已有处理,但需要调整什么参数?),谢谢! |
Beta Was this translation helpful? Give feedback.
-
我有的图片是90度旋转的,怎么判定啊,90度的图片识别效果很差 |
Beta Was this translation helpful? Give feedback.
-
PaddleOCR OCR analyzes Left-to-Right instead of Right-to-Left for Arabic- How to process RTL languages correctly? |
Beta Was this translation helpful? Give feedback.
-
再打扰一下 mac mini m4 安装完 paddle.utils.run_check() 检测只有一个cpu使用我看之前issue也有人提到过 |
Beta Was this translation helpful? Give feedback.
-
开启enable_hpi之后出现OOM,显存只有2G,请问大佬有解决办法嘛 |
Beta Was this translation helpful? Give feedback.
-
from pathlib import Path pipeline = PPStructureV3( For Imageoutput = pipeline.predict( 可视化结果并保存 json 结果for res in output: |
Beta Was this translation helpful? Give feedback.
-
只有PP-OCRv5 有该图片路径,其他的图片路径不存在。
|
Beta Was this translation helpful? Give feedback.
-
thanks for error on CPU version: E0608 16:30:35.157398 22224 analysis_config.cc:169] Please use PaddlePaddle with GPU version. |
Beta Was this translation helpful? Give feedback.
-
是否可以添加一个系统资源需求说明,例如我需要cpu部署或者gpu部署大概需要什么配置? |
Beta Was this translation helpful? Give feedback.
-
/opt/anaconda3/envs/ocr/lib/python3.8/site-packages/paddle/utils/cpp_extension/extension_utils.py:711: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md |
Beta Was this translation helpful? Give feedback.
-
gpu-cuda12.4 , cudnn9.1 有可用的paddle-gpu的版本么 谢谢!祝项目一路长虹! |
Beta Was this translation helpful? Give feedback.
-
对于生产使用 cpu ,直接 启用 高性能推理 安装 "paddleocr install_hpi_deps cpu" 后 把python的使用方式直接部署linux, 就可使用 pp-ocrv5了吧? 初始化: 提供端口: 启动: |
Beta Was this translation helpful? Give feedback.
-
C++ Traceback (most recent call last):No stack trace in paddle, may be caused by external reasons. Error Message Summary:FatalError: Segmentation fault (core dumped) 识别图片能正常工作,但用pdf文件就出现这个,有人知道是啥问题么 |
Beta Was this translation helpful? Give feedback.
-
use_tensorrt=True,使用tensorrt加速报错!! |
Beta Was this translation helpful? Give feedback.
-
你好,请问只能解析照片里的文字吗,直接解析pdf文件里的文字可以吗 |
Beta Was this translation helpful? Give feedback.
-
from paddleocr import PPStructureV3 TypeError: BasePredictor.init() got an unexpected keyword argument 'threshold' |
Beta Was this translation helpful? Give feedback.
-
怎么获取每一个字的坐标信息 |
Beta Was this translation helpful? Give feedback.
-
你好,我遇到一些问题, 安装依赖 ocr = PaddleOCR( ocr = PaddleOCR(use_doc_orientation_classify=True, use_doc_unwarping=True) # 文本图像预处理+文本检测+方向分类+文本识别ocr = PaddleOCR(use_doc_orientation_classify=False, use_doc_unwarping=False) # 文本检测+文本行方向分类+文本识别ocr = PaddleOCR(text_detection_model_name="PP-OCRv5_mobile_det",text_recognition_model_name="PP-OCRv5_mobile_rec",use_doc_orientation_classify=False,use_doc_unwarping=False,use_textline_orientation=False) # 更换 PP-OCRv5_mobile 模型result = ocr.predict("testImages\CarPlate\1.jpg") 这个是官网的这个命令,不过我把这个图片路径改成自己的了(是一个汽车车牌的),然后是联网的,模型问题不存在 |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
PaddleOCR/latest/quick_start
Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)
https://paddlepaddle.github.io/PaddleOCR/latest/quick_start.html
Beta Was this translation helpful? Give feedback.
All reactions