frotms

@jingwanli6666 在其他下载的一个det_r50_vd_east.tar是指在哪里下载的？如果是其他实现，要么需要修改参数的名字去适配此代码实现，要么修改此代码与其他实现一致。

`python tools/infer/predict_e2e.py --e2e_model_path ./en_server_pgnetA_infer.pth --image_dir ./doc/imgs_en/img623.jpg --e2e_algorithm PGNet --e2e_pgnet_polygon True --e2e_char_dict_path ./pytorchocr/utils/ic15_dict.txt --e2e_yaml_path ./configs/e2e/e2e_r50_vd_pg.yml` 请根据[推理文档](https://github.com/frotms/PaddleOCR2Pytorch/blob/main/doc/doc_ch/inference.md#%E7%AB%AF%E5%88%B0%E7%AB%AF%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86)推理，效果和PaddleOCR是一样的

用paddleocr训练即可

@mrlihellohorld @lindadamama @lianuo @pangmaoran 可以

@pangmaoran ``` # ch_PP-OCRv4_det python ./converter/ch_ppocr_v4_det_converter.py --yaml_path ./configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_student.yml --src_model_path ch_PP-OCRv4_det_train_dir # ch_PP-OCRv4_server_det python ./converter/ch_ppocr_v4_det_server_converter.py --yaml_path ./configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_teacher.yml --src_model_path ./models_v2.7/ppocrv4/ch_PP-OCRv4_det_server_train_dir # ch_PP-OCRv4_rec python ./converter/ch_ppocr_v4_rec_converter.py --yaml_path ./configs/rec/PP-OCRv4/ch_PP-OCRv4_rec.yml --src_model_path ch_PP-OCRv4_rec_train_dir # ch_PP-OCRv4_server_rec python ./converter/ch_ppocr_v4_rec_server_converter.py...

@zhangqiqi1228 需要用PaddleOCR的工程训练，将训好的模型再转成PyTorch

`python .\converter\ch_ppocr_v3_det_converter_student.py --src_model_path your_ch_PP-OCRv3_det_distill_train_dir` 使用如下文件 ``` import os, sys sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from collections import OrderedDict import numpy as np import cv2 import torch from pytorchocr.base_ocr_v20 import BaseOCRV20 class PPOCRv3DetConverter(BaseOCRV20): def...

> Hi, how to download a pre-trained model automatically? which argument do we have to use to download that data @hammadyounas2008 You may download a pretrained model manually. You can...

@hammadyounas2008 add it by yourself

`train.txt` is in flower_102.tar. There is no txt-generator here. Please DIY.