PaddleOCR icon indicating copy to clipboard operation
PaddleOCR copied to clipboard

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 de...

Results 1088 PaddleOCR issues
Sort by recently updated
recently updated
newest added

Hi Everyone, I trained the SVTR model on a 9M images dataset on a custom language using a custom dictionary. Then I exported the model into an inference model. When...

1.训练模型推理结果异常(infer_rec.py),推理模型推理正常(predict_rec),请问是什么原因导致?(已确认模型文件为官方下载,配置及字典文件严格对应) 通用识别模型推理:(采用官方训练模型,推理结果异常) python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml -o Global.pretrained_model=./best_accuracy Global.infer_img=test.jpg 繁体字模型推理:(采用官方训练模型,推理结果异常,已修改字典及配置文件) python3 tools/infer_rec.py -c configs/rec/PP-OCRv3/multi_language/chinese_cht_PP-OCRv3_rec.yml -o Global.pretrained_model=./best_accuracy Global.infer_img=./test.jpg --rec_char_dict_path="./ppocr/utils/dict/chinese_cht_dict.txt"

PaddleOCR是否仅支持申威服务器的Py推理,不支持C++ Paddle下只提供了申威的Py编译安装,暂无C++

直接调用官方代码训练后有记录train.log,如何直接绘制Loss、acc变化曲线

基于ch_ppocr_server_v2.0_rec_train直接转infer模型,再转onnx模型,在CPU上tps达到30,但是只要ch_ppocr_server_v2.0_rec_train经过一个step的训练之后,有了反向传播,在转infer模型,再转onnx之后的性能在CPU上只有5tps了,想问下官方这是什么原因。paddle版本2.0.0

- 系统环境/System Environment:Jetson AGX Orin (Jetpack 5.0) - Environment: Nvidia-docker 20.x, Python3.6, Cuda 11.4, based on nvcr.io/nvidia/deepstream-l4t:6.1.1-triton base image - Paddle: 2.3.0, installed from this wheel: https://paddle-inference-lib.bj.bcebos.com/2.3.0/python/Jetson/jetpack4.5_gcc7.5/all/paddlepaddle_gpu-2.3.0-cp36-cp36m-linux_aarch64.whl - PaddleOCR: tried...

- 想问下有没有针对这种情况的模型,或者有其他好的模型嘛 ![1668426727820](https://user-images.githubusercontent.com/40414529/201653578-ba1843b7-858f-477b-89c3-88b01b255666.png) ![1668426670320](https://user-images.githubusercontent.com/40414529/201653598-4d7a9d0a-0df4-4566-b458-a2f06ccc948b.png) 想问下有没有针对这种情况的模型,或者有其他好的模型嘛

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment:linux - 版本号/Version:2.6. Paddle: PaddleOCR:2.6 问题相关组件/Related components: ppstructure - 运行指令/Command Code: ```shell python paddleocr.py --type=structure --recovery=true --image_dir=./docs/table/1.png ```...