v3.0.3

• Bug修复：
  • 修复enable_mkldnn参数不生效的问题，恢复CPU默认使用MKL-DNN推理的行为。
  • 随PaddleX 3.0.3 版本的其他修复

v3.0.2

• 功能新增：

  • 模型默认下载源从BOS改为HuggingFace，同时也支持用户通过更改环境变量PADDLE_PDX_MODEL_SOURCE为BOS，将模型下载源设置为百度云对象存储BOS。
  • PP-OCRv5、PP-StructureV3、PP-ChatOCRv4等pipeline新增C++、Java、Go、C#、Node.js、PHP 6种语言的服务调用示例。
  • 优化PP-StructureV3产线中版面分区排序算法，对复杂竖版版面排序逻辑进行完善，进一步提升了复杂版面排序效果。
  • 优化模型选择逻辑，当指定语言、未指定模型版本时，自动选择支持该语言的最新版本的模型。
  • 为MKL-DNN缓存大小设置默认上界，防止缓存无限增长。同时，支持用户配置缓存容量。
  • 更新高性能推理默认配置，支持Paddle MKL-DNN加速。优化高性能推理自动配置逻辑，支持更智能的配置选择。
  • 调整默认设备获取逻辑，考虑环境中安装的Paddle框架对计算设备的实际支持情况，使程序行为更符合直觉。
  • 新增PP-OCRv5的Android端示例，详情。

• Bug修复：

  • 修复PP-StructureV3部分CLI参数不生效的问题。
  • 修复部分情况下export_paddlex_config_to_yaml无法正常工作的问题。
  • 修复save_path实际行为与文档描述不符的问题。
  • 修复基础服务化部署在使用MKL-DNN时可能出现的多线程错误。
  • 修复Latex-OCR模型的图像预处理的通道顺序错误。
  • 修复文本识别模块保存可视化图像的通道顺序错误。
  • 修复PP-StructureV3中表格可视化结果通道顺序错误。
  • 修复PP-StructureV3产线中极特殊的情况下，计算overlap_ratio时，变量溢出问题。

• 文档优化：

  • 更新文档中对enable_mkldnn参数的说明，使其更准确地描述程序的实际行为。
  • 修复文档中对lang和ocr_version参数描述的错误。
  • 补充通过CLI导出产线配置文件的说明。
  • 修复PP-OCRv5性能数据表格中的列缺失问题。
  • 润色PP-StructureV3在不同配置下的benchmark指标。

• 其他：

  • 放松numpy、pandas等依赖的版本限制，恢复对Python 3.12的支持。

v3.0.1

• 优化部分模型和模型配置：
  • 更新 PP-OCRv5默认模型配置，检测和识别均由mobile改为server模型。为了改善大多数的场景默认效果，配置中的参数limit_side_len由736改为64
  • 新增文本行方向分类PP-LCNet_x1_0_textline_ori模型，精度99.42%，OCR、PP-StructureV3、PP-ChatOCRv4产线的默认文本行方向分类器改为该模型
  • 优化文本行方向分类PP-LCNet_x0_25_textline_ori模型，精度提升3.3个百分点，当前精度98.85%
• 优化3.0.0版本部分存在的问题
  • 优化CLI使用体验： 当使用PaddleOCR CLI不传入任何参数时，给出用法提示。
  • 新增参数： PP-ChatOCRv3、PP-StructureV3支持use_textline_orientation参数。
  • CPU推理速度优化： 所有产线CPU推理默认开启MKL-DNN。
  • C++推理支持： PP-OCRv5的检测和识别串联部分支持C++推理
• 修复3.0.0版本部分存在的问题
  • 修复由于公式识别、表格识别模型无法使用MKL-DNN导致PP-StructureV3在部分cpu推理报错的问题
  • 修复在部分GPU环境中推理报FatalError: Process abort signal is detected by the operating system错误的问题
  • 修复部分Python3.8环境的type hint的问题
  • 修复PPStructureV3.concatenate_markdown_pages方法不存在的问题。
  • 修复实例化paddleocr.PaddleOCR时同时指定lang和model_name时model_name不生效的问题。

v3.0.0

• 发布全场景文字识别模型PP-OCRv5： 单模型支持五种文字类型和复杂手写体识别；整体识别精度相比上一代提升13个百分点。

• 发布通用文档解析方案PP-StructureV3： 支持多场景、多版式 PDF 高精度解析，在公开评测集中领先众多开源和闭源方案。

• 发布智能文档理解方案PP-ChatOCRv4： 原生支持文心大模型4.5 Turbo，精度相比上一代提升15个百分点。

• 重构部署能力，统一推理接口： PaddleOCR 3.0 融合了飞桨 PaddleX3.0 工具的底层能力，全面升级推理、部署模块，优化 2.x 版本的设计，统一并优化了 Python API 和命令行接口（CLI）。部署能力现覆盖高性能推理、服务化部署及端侧部署三大场景。

• 适配飞桨框架 3.0，优化训练流程： 新版本已兼容飞桨 3.0 的 CINN 编译器等最新特性，静态图模型存储文件名由 xxx.pdmodel 改为 xxx.json。

• 统一模型名称： 对PaddleOCR3.0支持的模型命名体系进行了更新，采用更规范、统一的命名规则，为后续迭代与维护奠定基础。

v2.10.0

What's Changed

• update docs by @cuicheng01 in #14031
• update paddle2onnx doc by @inisis in #14038
• fix gpu memory growth by @zhangyubo0722 in #14037
• updata en docs by @dyning in #14036
• fix nan in PP-OCRv4 by @wangna11BD in #14043
• update a live promotion by @Zhiiixin in #14042
• reset latex ocr by @zhangyubo0722 in #14046
• Update pyproject.toml for add dependency by @Liyulingyue in #14058
• Fix CMAKE_CXX_FLAGS optimize flag by @Hirozy in #14059
• fix isnan_v2 is not supported in paddle2onnx by @GreatV in #14060
• ci: Fixed docs multi version error by @SWHL in #14048
• fix hyperlinks by @AmberC0209 in #14073
• fix nan in ppocrv4 for benchmark by @wangna11BD in #14072
• ci: Support seperate update of branch docs by @SWHL in #14079
• ci: fixed main doc ci by @SWHL in #14084
• Allow create_predictor function to accept array of ONNX Execution Providers by @Salmondx in #14078
• docs: update quickstart by @SWHL in #14108
• docs: add command line usage documentation of quickstart page by @SWHL in #14110
• docs: add installation documentation of paddle by @SWHL in #14117
• docs: fixed typo by @SWHL in #14118
• image without any text will show a warning by @GreatV in #14132
• doc: remove duplicate paragraphs by @GreatV in #14133
• docs: update paddle2onnx documentations by @GreatV in #14144
• [third-party] Fix the issue of inference errors with KIE mode in ONNX format by @Alex37882388 in #14138
• update tests PR CI github action by @GreatV in #14159
• 移除doc目录下文档，保留fonts和doc_i18n两个目录 by @SWHL in #14156
• 移除ppstructure目录下旧有文档 by @SWHL in #14161
• docs: fixed error image link (#14164) by @SWHL in #14165
• 更新i18n的首页内容到新站点 by @SWHL in #14166
• docs: fix i18n languange code error by @SWHL in #14167
• docs: fix syntax error by @SWHL in #14168
• docs: update i18n docs by @GreatV in #14169
• upgrade to numpy 2.0 and remove imgaug by @GreatV in #13937
• docs: format multi languange docs home page by @SWHL in #14170
• docs: add the missing image by @GreatV in #14180
• Create close_inactive_issues.yaml by @GreatV in #14183
• update hpi config by @zhangyubo0722 in #14076
• Update close_inactive_issues.yaml by @GreatV in #14189
• Update close_inactive_issues.yaml by @GreatV in #14190
• remove lock inactive issues by @GreatV in #14192
• fix benchmark bug by @changdazhou in #14194
• pre-commit autoupdate && pre-commit run --all-files by @cclauss in #14201
• Remove Python 2 compatibility dependency six by @cclauss in #14202
• update quick_start by @AmberC0209 in #14200
• rename train result by @zhangyubo0722 in #14217
• fix benchmark bug by @changdazhou in #14235
• fix benchmark det_r50_vd_pse_v2_0 train error by @GreatV in #14239
• update infer/utility.py to support json format model by @GreatV in #14233
• Support inference for GCU by @EnflameGCU in #14142
• update docs by @AmberC0209 in #14230
• fix: Title text partially missing issue in recovery_to_markdown.py by @Coobiw in #14216
• change_support list by @liuhongen1234567 in #14293
• support latexocr static train by @liuhongen1234567 in #14297
• docs: Fix chinese image being displayed on the english readme page by @khanfarhan10 in #14299
• docs: update quick_start and recognition doc by @GreatV in #14302
• add d2s_train_image_shape for static train by @liuhongen1234567 in #14312
• update install command by @AmberC0209 in #14314
• fix: unable to export images without text to docx format by @GreatV in #14306
• paddle.shape return int64 tensor by @wanghuancoder in #14318
• docs: add warning of Apolications part by @SWHL in #14338
• Update algorithm_rec_cppd.md by @GreatV in #14366
• Update 印章弯曲文字识别.md by @BUJIQI in #14368
• update_det_static by @Sunting78 in #14372
• fix:calcute the left_center_pt and right_center_pt from min_area_quad by @fangfangzk in #14363
• add unimernet model by @liuhongen1234567 in #14357
• fix shape64 by @wanghuancoder in #14376
• add slanext models by @liu-jiaxuan in #14374
• fix: replace rec_image_shape when manually set by @JesuisTong in #14371
• repair type bug for ppocrv3 by @liuhongen1234567 in #14397
• [WIP]support export with pir and no pir by @zhangyubo0722 in #14379
• Add pp formulanet by @liuhongen1234567 in #14429
• repair formula bug when export by @liuhongen1234567 in #14442
• modify export with pir by @zhangyubo0722 in #14441
• update SLANet inference weights for adapt to paddle3.0b2 by @cuicheng01 in #14467
• fix_server_v4_det_output by @Sunting78 in #14472
• fix label_dict save bug by @zhangyubo0722 in #14273
• add ppocrv4_doc dict by @liuhongen1234567 in #14499
• fix latex_ocr inference by @vivienfanghuagood in #14498
• fix SLANeXt export bug by @liu-jiaxuan in #14512
• add version control for export and modify hpi config by @zhangyubo0722 in #14513
• fix slanext export bug by @liu-jiaxuan in #14519
• repair bug in latexocr cpu infer and typo in bleu score by @liuhongen1234567 in #14552
• Fix language error and spelling mistakes in the documentation by @timminator in #14571
• Keep GitHub Actions up to date with GitHub's Dependabot by @cclauss in #14569
• repair train bug in multi gpu by @liuhongen1234567 in #14576
• build(deps): bump the github-actions group with 3 updates by @dependabot in #14573
• remove max inplace grad by @phlrain in #14596
• build(deps): bump pypa/gh-action-pypi-publish from 1.12.3 to 1.12.4 in the github-actions group by @dependabot in #14603
• CPP: emplace_back() replaces many push_back()...to improve performance by @nonwill in #14610
• Add Thai character dictionary for OCR recognition by @Thanajade in #14620
• CPP: Make functions mostly noexcept to improve runtime performance by @nonwill in #14613
• CPP: tidied file header includes by @nonwill in #14621
  *...

v2.9.1

What's Changed

• [cherry-pick] update paddle2onnx doc by @inisis in #14051
• [cherry-pick] fix isnan is not supported and update pyproject by @GreatV in #14061
• [cherry-pick] ci: Fixed docs multi version error by @GreatV in #14063

Full Changelog: v2.9.0...v2.9.1

v2.9.0

What's Changed

• fix: table recognition content is not escaped properly by @GreatV in #13277
• fix bug when layout_predictor is None by @GreatV in #13279
• add url in pyproject, and update version number by @GreatV in #13274
• unifying data types in the SLAHead by @GreatV in #13276
• add PaddleX info to README by @TingquanGao in #13308
• Update expired link in quickstart.md by @ZeddYu in #13253
• optimize func: get_infer_gpuid by @GreatV in #13275
• fix slice op parameters not being passed correctly by @GreatV in #13319
• Solve ModuleNotFoundError: No module named 'tools.infer' by @myhloli in #13348
• Add hardware docs by @nepeplwu in #13329
• add paddlex link by @TingquanGao in #13316
• Fix the dictionary bug in tablerec inference by @Topdu in #13362
• add bn_dict.txt by @taeefnajib in #13373
• add missing docstring in paddleocr.py using copilot by @jzhang533 in #13344
• line 445 program.py by @ManikSinghSarmaal in #13389
• fix layout recovery import error by @GreatV in #13434
• Latexocr paddle by @liuhongen1234567 in #13401
• [doc]add amp train notes for detection train by @andyjiang1116 in #13481
• remove some of the less common dependencies by @GreatV in #13461
• docs: Add a new document site by @SWHL in #13375
• Update mkdocs.yml by @GreatV in #13487
• chore: Update issue template by @SWHL in #13505
• chore: Update bug report template by @SWHL in #13508
• Fix cpp_infer "--enable_mkldnn=false" not effective by @hiroi-sora in #13539
• Enable Main Branch Support for PaddleX by @zhangyubo0722 in #13523
• docs: Update README by @SWHL in #13543
• docs: Update README_en by @SWHL in #13545
• 修改错别字 by @MonkeyBrothers in #13544
• docs: Remove old applications docs by @SWHL in #13551
• fix: 'numpy' has no attribute 'astype' by @laolitou in #13554
• add latexocr docs and fix some typos by @GreatV in #13532
• chore(Issue_template): Add validation of Environment and MPE code by @SWHL in #13559
• skip text files when running test ci by @GreatV in #13561
• fix bug for paddlepaddle3.0 by @changdazhou in #13568
• docs: Update the pdf file path in the operation demonstration by @Gmgge in #13575
• support benchmark for paddlepaddle3.0 by @changdazhou in #13574
• improve the reading experience of some documents by @GreatV in #13562
• update dive into OCR book link by @GreatV in #13581
• docs: Shorten the image path and remove dupliate images by @SWHL in #13585
• docs: Fix docs errors by @SWHL in #13588
• skip text files when running test ci on push by @GreatV in #13582
• docs: Add android_demo docs by @SWHL in #13601
• fix download bug when use multi gpus by @changdazhou in #13610
• disable automatic checks for new version albumentations by @GreatV in #13583
• 修复LaTeXOCR 在paddleX中的一些问题 by @liuhongen1234567 in #13646
• update docs and remove out-of-date event by @GreatV in #13660
• setuptools 72.2.0 result in that MANIFEST.in is invalid by @TingquanGao in #13670
• update docs and remove old docs by @GreatV in #13662
• update docs and fix markdown render error by @GreatV in #13678
• chore: Update issue template by @SWHL in #13679
• cache Python dependencies and PaddleOCR files by @GreatV in #13682
• Add files via upload by @lingskr in #13685
• Update ch_PP-OCRv4_rec_distillation.yml by @jiqirenfeile in #13692
• Remove channel links from docs by @zhangyubo0722 in #13674
• Code Style Unification by @zhangyubo0722 in #13697
• docs: Remove doc/datasets directory and fix docs/datasets documents by @SWHL in #13700
• Provides Vietnamese dictionary and corpus by @lingskr in #13698
• Modify the data processing part of LaTeXOCR and replace the absolute path by a relative path by @liuhongen1234567 in #13702
• use setuptools-scm extracts PaddleOCR versions by @GreatV in #13716
• Repair the bug in the inference script for LaTeX OCR by @liuhongen1234567 in #13750
• fixed: mkldnn -> onednn by @achieve-dream1221 in #13757
• remove unused enumerate by @Kayzwer in #13760
• update applications/overview.md by @GreatV in #13763
• Fix setting of make border epoch by @Sunting78 in #13783
• Fix doc link in docs by @Topdu in #13792
• Add support for Hebrew Language and Alphabet by @johnlockejrr in #13797
• Add Syriac script support by @johnlockejrr in #13800
• update KIE docs by @GreatV in #13799
• fix the CI running errors in tests. by @GreatV in #13846
• Fix pir dy2st train by @0x45f in #13853
• fix SRN algorithm infer error by @GreatV in #13851
• update pretrain for benchmark by @changdazhou in #13820
• fix bugs for SLANet infer by @liu-jiaxuan in #13861
• fix version by @TingquanGao in #13895
• set --image_dir to be required by @GreatV in #13896
• support export after save model by @zhangyubo0722 in #13844
• fix hubserving run error by @GreatV in #13918
• fix lateocr bug by @zhangyubo0722 in #13920
• 1.在ppstructure管道中添加latex_ocr公式识别功能；2.添加pdf转markdown文件功能 by @ztyf-lq in #13868
• updata 2.9, adding new models and supporting all-in-one full developm… by @dyning in #13932
• updata 2.9, adding new models and supporting all-in-one full developm… by @dyning in #13933
• adding new models and supporting all-in-one full development tools by @dyning in #13934
• Update quick_start.md with html, not md by @dyning in #13935
• Update quick_start.md for paddlex by @dyning in #13936
• pdf to markdown document by @ztyf-lq in #13942
• Update algorithm_rec_vitstr_en.md by @GreatV in #13947
• update a live promotion by @Zhiiixin in #13954
• ci: Support multi version docs by @SWHL in #13957
• docs: Add tip of old documents by @SWHL in #13960
• ci: Fix mike error by @SWHL in #13962
• Update README.md, fixed broken quick start link by @Kozmosa in #13965
• fix broken link by @GreatV in #13970
• [NPU] cherrypick13983 by @Wang...

v2.8.1

What's Changed

• [cherry-pick] add project url and fix a bug by @GreatV in #13281
• [cherry-pick] fix slice op parameters not being passed correctly (#13319) by @GreatV in #13324
• Fix the dictionary bug in tablerec inference by @Topdu in #13364

Full Changelog: v2.8.0...v2.8.1

v2.8.0

终于等到你！PaddleOCR 新版本发布！

What's Changed

• [Cherry-pick] #10515 by @ToddBear in #10537
• [BugFix]compat_pillow by @shiyutang in #10596
• [bug fix] fix none res in recovery by @andyjiang1116 in #10603
• Fix seed passing issue of build_dataloader by @RuohengMa in #10614
• [bug fix]rm invalid params by @andyjiang1116 in #10605
• [Cherry-pick] #10441 #10512 by @moehuster in #10593
• 修改数据增强导致的DSR报错 by @xu-peng-7 in #10662
• onnxruntime support gpu by @WenmuZhou in #10668
• Update VQA to use the updated LayoutLM syntax from PaddleNLP by @sijunhe in #9791
• 实现功能：当--savefile为true时，在--output下以当前图片名称后接“.txt”为文件名保存ocr推理结果，解决了issues： by @WilliamQf-AI in #10628
• Cherrypicking GH-10217 and GH-10216 to PaddlePaddle:dygraph by @UserUnknownFactor in #10654
• fix numpy speed by @wanghuancoder in #10773
• Cherrypicking GH-10251 & GH-10181 to PaddleOCR:dygraph by @itasli in #10710
• rec_r45_abinet.yml add max_length and image_size by @xlg-go in #10744
• ch_PP-OCRv4_rec_distill.yml, fix KeyError: 'NRTRLabelDecode' by @xlg-go in #10761
• 根据推理对三通道的图像需求，以及opencv中imread参数说明IMREAD_COLOR(If set, always convert … by @Gmgge in #10777
• Update algorithm_kie_vi_layoutxlm_en.md by @sagarjgb in #10736
• Add new recognition method "ParseQ" by @ToddBear in #10836
• rm fluid for paddle dev by @tink2123 in #10931
• rec_r45_abinet for export model by @xlg-go in #10892
• fix:修复通道数不匹配造成的PPOCRLabel启动失败问题#10748,根据更新日志发现#10655，由于paddleocr中增加了对… by @Gmgge in #10847
• [New] add rec CPPD model by @Topdu in #10990
• fix cls_x and bbox_x is possibly unbound by @SigureMo in #10991
• add svtr large model by @zhangyubo0722 in #10937
• [WIP]support eval pre epoch by @zhangyubo0722 in #11003
• Update kie_datasets_en.md by @sagarjgb in #10735
• fix import collection for py310 by @tink2123 in #11012
• update ppocrv4_framework by @tink2123 in #11048
• Update how_to_do_kie_en.md by @sagarjgb in #10731
• add cppd u14m train model and doc by @Topdu in #11052
• Fixed bug with "max_text_length" for VisionLAN by @victor30608 in #11025
• Cherrypicking GH-10923 to PaddleOCR:dygraph by @itasli in #11069
• Update quickstart_en.md by @sagarjgb in #10732
• Update README.md by @sagarjgb in #10733
• Update algorithm_overview_en.md by @sagarjgb in #10734
• [Cherry-pick] Cherry-pick from release/2.6 by @shiyutang in #11092
• [TIPC]update tipc scripts by @USTCKAY in #11097
• fix satrn export for paddle2.5 by @tink2123 in #11096
• [BugFix]Fix parseq net by @shiyutang in #11126
• update uygur dict by @hfengzhi in #11125
• Add tipc for "ParseQ" method by @ToddBear in #10843
• fix SAR inference, when batch size>1, norm_img_batch and valid_ratios… by @shiyunalex in #11238
• v4 det cml configs by @sylarwcy in #11258
• 解決recognition的train test分割程式執行後的文檔每行間多出一行空格 by @DingHsun in #11280
• Fix for Ambiguous Boolean Evaluation Error in PaddleOCR with Python 3.11 by @muhammadAgfian96 in #11287
• Dygraph【benchmark】add max_mem_reserved for benchmark by @mmglove in #11284
• Fix bug when running on XPU by @RuohengMa in #11299
• Dygraph by @RuohengMa in #11301
• Dygraph fix max_mem_reserved for benchmark by @mmglove in #11341
• 在check_gpu时增加对当前环境可用设备的检查 by @TracebaK in #11293
• Fixed some bugs that caused PPOCRLabel to crash, added ability to expand checkboxes by @g39088902 in #11236
• fix a bug for rec_postprocess.py by @Ataraxy33 in #11389
• Optimize prediction on long image and deduplicate similar boxes with multiple lables by @marswen in #11366
• doc: add doc for satrn by @wkml in #11397
• Update zeros' comment in rec_abinet_head.py by @YesianRohn in #11374
• Fix QPointF IndexError: list index out of range by @firmament2008 in #11393
• update paddlex of readme by @zhangyubo0722 in #11422
• chore: add notes for docker gpu deploy PP-OCRv4 by @sheiy in #11390
• Fix words by @co63oc in #11448
• [Feature]Complete the ppocrv4_act by @ranchongzhi in #11345
• rm QR code in the document by @tink2123 in #11512
• rm QR code by @tink2123 in #11532
• Fix dead links by @MatKollar in #11520
• cherry-pick for lazy import pymupdf and pre-commit by @tink2123 in #11692
• adapter new type promotion rule for Paddle 2.6 by @zxcd in #11698
• setup a workflow for publishing package to pypi by @jzhang533 in #11804
• update link mentioned at #11763 by @jzhang533 in #11764
• fix AttributeError by @GreatV in #11686
• fix: Correct misuse of try_import from paddle.utils by @neteroster in #11820
• Update quickstart.md for a better python pdf demo by @qwedc001 in #11927
• Update quickstart_en.md by @qwedc001 in #11934
• Enhance the OCR recognition accuracy of PPStructure. by @RussellLuo in #11916
• add u14m results of cppd by @Topdu in #11943
• use tensor.shape bug not paddle.shape(tensor) by @wanghuancoder in #11919
• add pre-commit workflow by @GreatV in #11973
• docs: Update FAQ.md, delete repeated question by @xu8117 in #11972
• Fix the bug where Python scripts fail to execute PDF text recognition… by @guangyunms in #11994
• 【OCR Issue No.9】以可选形式支持Visualdl by @Liyulingyue in #11947
• fix weird version info by @GreatV in #12003
• 【OCR Issue No.9】移除明确不适合放在ppocr依赖中的依赖项 by @Liyulingyue in #11946
• Burmese Language dict and corpus by @1chimaruGin in #12020
• 面版识别添加onnx支持完善 by @heweisheng in #12068
• Update README.md by @dyning in #12086
• fix readme codestyle by @GreatV in #12095
• fix wrong link for 通用OCR in README.txt by @tackhwa in #12100
• move PPOCRLabel to PFCCLab/PPOCRLabel by @GreatV in #12104
• move StyleText to PFCCLab/StyleText by @GreatV in #12121
• openocr compti code by @Topdu in #12033
• table rec code by @invictuszhao in #11999
• Error with pyclipper inhomogeneous expanded array by @zovelsanj in #12108
• 【OCR Issue No.2】修复训练过程中找不到對應模型和训练时计算精度报错 by @mattheliu in https://github.com/PaddlePaddle/Paddle...

PaddleOCRv2.7.5

fix broken v2.7.4