目錄1. 新建文件夾2. 準(zhǔn)備數(shù)據(jù)3. 數(shù)據(jù)集的劃分注意1注意24. 下載預(yù)訓(xùn)練模型5. 訓(xùn)練文字檢測模型6. 訓(xùn)練文字識別模型7. 測試8. 轉(zhuǎn)換為推理模型9. 檢測模型和識別模型推理1. 新建文件夾進(jìn)入PPOCRLabel源碼目錄在上一層目錄新建文件夾train_data在train_data中新建文件夾drivingData。2. 準(zhǔn)備數(shù)據(jù)在PPOCRLabel中標(biāo)注好的數(shù)據(jù)如下圖將標(biāo)注好的數(shù)據(jù)拷貝到drivingData文件夾。3. 數(shù)據(jù)集的劃分cd到PPOCRLabel源碼路徑運(yùn)行劃分?jǐn)?shù)據(jù)集代碼python gen_ocr_train_val_test.py--trainValTestRatio6:2:2--datasetRootPath../train_data/drivingData會在train_data文件夾下出現(xiàn)如下文件夾可刪除drivingData文件夾。注意1新版本gen_ocr_train_val_test.py中做了操作系統(tǒng)兼容處理在linux下不會報錯代碼如下部分代碼withopen(label_file_path,r,encodingUTF-8)aslabel_file:label_file_contentlabel_file.readlines()random.shuffle(label_file_content)label_record_lenlen(label_file_content)forindex,label_record_infoinenumerate(label_file_content):image_relative_path,image_labellabel_record_info.split(\t)image_nameos.path.basename(image_relative_path)ifflagdet:image_pathos.path.join(data_abs_path,image_name)elifflagrec:image_pathos.path.join(data_abs_path,args.recImageDirName,image_name)老版本gen_ocr_train_val_test.py中l(wèi)abelFileReadopen(labelFilePath,r,encodingUTF-8)labelFileContentlabelFileRead.readlines()random.shuffle(labelFileContent)labelRecordLenlen(labelFileContent)forindex,labelRecordInfoinenumerate(labelFileContent):imageRelativePathlabelRecordInfo.split(\t)[0]imageLabellabelRecordInfo.split(\t)[1]imageNameos.path.basename(imageRelativePath)ifflagdet:imagePathos.path.join(dataAbsPath,imageName)elifflagrec:imagePathos.path.join(dataAbsPath,{}\\{}.format(args.recImageDirName,imageName))在windows下使用以上代碼在linux下將上面代碼最后一行的“{}\{}”改為“{}/{}”。注意2如果是在云端AutoDL上進(jìn)行訓(xùn)練建議在云端linux進(jìn)行上面的數(shù)據(jù)劃分代碼如果在windows下劃分的數(shù)據(jù)集路徑為如下格式訓(xùn)練要在云端需要改路徑D:\項目\train_data\det\train\2026_6_18_11_22_25_Digital.jpg[{transcription:23.6,points:[[287,152],[509,152],[509,277],[287,277]],difficult:false},{transcription:89,points:[[369,271],[505,271],[505,401],[369,401]],difficult:false}]D:\項目\train_data\det\train\2026_6_16_16_25_30_Digital.jpg[{transcription:21.5,points:[[300,197],[480,197],[480,283],[300,283]],difficult:false},{transcription:41.9,points:[[280,278],[485,278],[485,372],[280,372]],difficult:false}]D:\項目\train_data\det\train\2026_6_17_15_29_20_Digital.jpg[{transcription:20.2,points:[[305,154],[507,154],[507,280],[305,280]],difficult:false},{transcription:56,points:[[373,271],[502,271],[502,400],[373,400]],difficult:false}]train.txt文件中D:\項目\train_data\det\train替換/root/PaddleOCR/PaddleOCR-main/train_data/det/train/val.txt替換/root/PaddleOCR/PaddleOCR-main/train_data/det/val/test.txt替換/root/PaddleOCR/PaddleOCR-main/train_data/det/test/#rec文件夾也類似替換。4. 下載預(yù)訓(xùn)練模型模型列表將下載的預(yù)訓(xùn)練模型放在如下文件夾下5. 訓(xùn)練文字檢測模型打開如下文件做如下修改Global: model_name: PP-OCRv5_server_det# To use static model for inference.debug:falseuse_gpu:trueepoch_num:epoch_num500log_smooth_window:20print_batch_step:10save_model_dir: /root/autodl-fs/output-1/PP-OCRv5_server_det save_epoch_step:10eval_batch_step: -0-200cal_metric_during_train:falsecheckpoints: pretrained_model: /root/PaddleOCR/PaddleOCR-main/pretrain_models/PP-OCRv5_server_det_pretrained.pdparams save_inference_dir: null use_visualdl:falseinfer_img: doc/imgs_en/img_10.jpg save_res_path: ./checkpoints/det_db/predicts_db.txt distributed:trueArchitecture: model_type: det algorithm: DB Transform: null Backbone: name: PPHGNetV2_B4 det: True Neck: name: LKPAN out_channels:256intracl:trueHead: name: PFHeadLocal k:50mode:largeLoss: name: DBLoss balance_loss:truemain_loss_type: DiceLoss alpha:5beta:10ohem_ratio:3Optimizer: name: Adam beta1:0.9beta2:0.999lr: name: Cosine learning_rate:0.001#(8*8c)warmup_epoch:2regularizer: name: L2 factor: 1e-6 PostProcess: name: DBPostProcess thresh:0.3box_thresh:0.6max_candidates:1000unclip_ratio:1.5Metric: name: DetMetric main_indicator: hmean Train: dataset: name: SimpleDataSet data_dir: /root/PaddleOCR/PaddleOCR-main/train_data label_file_list: - /root/PaddleOCR/PaddleOCR-main/train_data/det/train.txt ratio_list:[1.0]transforms: - DecodeImage: img_mode: BGR channel_first:false- DetLabelEncode: null - CopyPaste: null - IaaAugment: augmenter_args: - type: Fliplr args: p:0.5- type: Affine args: rotate: --10-10- type: Resize args: size: -0.5-3- EastRandomCropData: size: -640-640max_tries:50keep_ratio:true- MakeBorderMap: shrink_ratio:0.4thresh_min:0.3thresh_max:0.7total_epoch: *epoch_num - MakeShrinkMap: shrink_ratio:0.4min_text_size:8total_epoch: *epoch_num - NormalizeImage: scale:1./255. mean: -0.485-0.456-0.406std: -0.229-0.224-0.225order: hwc - ToCHWImage: null - KeepKeys: keep_keys: - image - threshold_map - threshold_mask - shrink_map - shrink_mask loader: shuffle:truedrop_last:falsebatch_size_per_card:8num_workers:8Eval: dataset: name: SimpleDataSet data_dir: /root/PaddleOCR/PaddleOCR-main/train_data label_file_list: - /root/PaddleOCR/PaddleOCR-main/train_data/det/val.txt transforms: - DecodeImage: img_mode: BGR channel_first:false- DetLabelEncode: null - DetResizeForTest: - NormalizeImage: scale:1./255. mean: -0.485-0.456-0.406std: -0.229-0.224-0.225order: hwc - ToCHWImage: null - KeepKeys: keep_keys: - image - shape - polys - ignore_tags loader: shuffle:falsedrop_last:falsebatch_size_per_card:4num_workers:8profiler_options: null打開anaconda激活虛擬環(huán)境進(jìn)入到PaddleOCR-Main路徑下輸入下列代碼開始訓(xùn)練python tools/train.py-c configs/det/PP-OCRv5/PP-OCRv5_server_det.yml6. 訓(xùn)練文字識別模型做如下修改Global: model_name: PP-OCRv5_server_rec# To use static model for inference.debug:falseuse_gpu:trueepoch_num:500log_smooth_window:20print_batch_step:10save_model_dir: /root/autodl-fs/output-1/PP-OCRv5_server_rec save_epoch_step:10eval_batch_step:[0,200]cal_metric_during_train:truecalc_epoch_interval:1pretrained_model: /root/PaddleOCR/PaddleOCR-main/pretrain_models/PP-OCRv5_server_rec_pretrained.pdparams checkpoints: save_inference_dir: use_visualdl:falseinfer_img: doc/imgs_words/ch/word_1.jpg character_dict_path: ./ppocr/utils/dict/weily.txt max_text_length:max_text_length25infer_mode:falseuse_space_char:truedistributed:truesave_res_path: /root/autodl-fs/output/rec/predicts_ppocrv5.txt d2s_train_image_shape:[3,48,320]Optimizer: name: Adam beta1:0.9beta2:0.999lr: name: Cosine learning_rate:0.0005warmup_epoch:1regularizer: name: L2 factor:3.0e-05 Architecture: model_type: rec algorithm: SVTR_HGNet Transform: Backbone: name: PPHGNetV2_B4 text_rec: True Head: name: MultiHead head_list: - CTCHead: Neck: name: svtr dims:120depth:2hidden_dims:120kernel_size:[1,3]use_guide: True Head: fc_decay:0.00001- NRTRHead: nrtr_dim:384max_text_length: *max_text_length Loss: name: MultiLoss loss_config_list: - CTCLoss: - NRTRLoss: PostProcess: name: CTCLabelDecode Metric: name: RecMetric main_indicator: acc Train: dataset: name: MultiScaleDataSet ds_width:falsedata_dir: /root/PaddleOCR/PaddleOCR-main/train_data ext_op_transform_idx:1label_file_list: - /root/PaddleOCR/PaddleOCR-main/train_data/rec/train.txt transforms: - DecodeImage: img_mode: BGR channel_first:false- RecAug: - MultiLabelEncode: gtc_encode: NRTRLabelEncode - KeepKeys: keep_keys: - image - label_ctc - label_gtc - length - valid_ratio sampler: name: MultiScaleSampler scales:[[320,32],[320,48],[320,64]]first_bs:bs128fix_bs:falsedivided_factor:[8,16]# w, his_training: True loader: shuffle:truebatch_size_per_card: *bs drop_last:truenum_workers:16Eval: dataset: name: SimpleDataSet data_dir: /root/PaddleOCR/PaddleOCR-main/train_data label_file_list: - /root/PaddleOCR/PaddleOCR-main/train_data/rec/val.txt transforms: - DecodeImage: img_mode: BGR channel_first:false- MultiLabelEncode: gtc_encode: NRTRLabelEncode - RecResizeImg: image_shape:[3,48,320]- KeepKeys: keep_keys: - image - label_ctc - label_gtc - length - valid_ratio loader: shuffle:falsedrop_last:falsebatch_size_per_card:128num_workers:4weily.txt文件如下0123456789-.輸入下列代碼開始訓(xùn)練python tools/train.py-c configs/rec/PP-OCRv5/PP-OCRv5_server_rec.yml7. 測試文本檢測模型測試python tools/infer_det.py-c configs/det/PP-OCRv5/PP-OCRv5_server_det.yml-o Global.pretrained_model./output/PP-OCRv5_det/best_accuracy.pdparams Global.infer_imgC:\Users\langr\Desktop\PaddleOCR\PaddleOCR-main\test.jpg文本識別模型測試python tools/infer_rec.py-c configs/rec/PP-OCRv5/PP-OCRv5_server_rec.yml-o Global.pretrained_model./output/PP-OCRv5_rec/best_accuracy.pdparams Global.infer_imgC:\Users\langr\Desktop\PaddleOCR\PaddleOCR-main\test_crop.jpg8. 轉(zhuǎn)換為推理模型文本檢測模型python tools/export_model.py-c./configs/det/PP-OCRv5/PP-OCRv5_server_det.yml-o Global.pretrained_model./output/PP-OCRv5_server_det/best_accuracy.pdparamsGlobal.save_inference_dir./inference_model/det/文本識別模型python tools/export_model.py-c./configs/rec/PP-OCRv5/PP-OCRv5_server_rec.yml-o Global.pretrained_model./output/PP-OCRv5_server_rec/latest.pdparamsGlobal.save_inference_dir./inference_model/rec/9. 檢測模型和識別模型推理步驟8的權(quán)重放在inference_model中測試代碼如下frompaddleocrimportPaddleOCR,TextDetection,TextRecognition modelPaddleOCR(det_model_dirr/root/PaddleOCR/PaddleOCR-main/inference_model/det,rec_model_dirr/root/PaddleOCR/PaddleOCR-main/inference_model/rec,use_doc_orientation_classifyFalse,use_doc_unwarpingFalse,use_textline_orientationFalse,text_det_unclip_ratio1.5,text_det_limit_side_len1580,# 1080*2text_det_limit_typemax,text_det_thresh0.3,text_det_box_thresh0.6)outputmodel.predict(r/root/PaddleOCR/PaddleOCR-main/train_data/det/test/2026_8_6_10_44_04_Digital.jpg)forresinoutput:res.save_to_img(save_path./output/)res.save_to_json(save_path./output/res.json)