317eac13b5
Add device check in check_gpu ( #11293 )
2023-12-06 11:12:58 +08:00
856ce76be3
Dygraph fix max_mem_reserved for benchmark ( #11341 )
...
* fix profile
* fix python3.10
* add max_mem_reserved for benchmark
* fix benchmark
2023-12-05 14:55:38 +08:00
dc001ac44a
Update utility.py
2023-11-30 12:32:17 +08:00
317f56a3f4
Refine according to comment ( #11301 )
2023-11-24 16:50:34 +08:00
c8544d04ec
Fix bug when running on XPU ( #11299 )
2023-11-23 16:58:48 +08:00
a0901d2175
Dygraph【benchmark】add max_mem_reserved for benchmark ( #11284 )
...
* fix profile
* fix python3.10
* add max_mem_reserved for benchmark
2023-11-22 20:11:14 +08:00
42d280907a
Fix ValueError in text detection due to ambiguous array evaluation ( #11287 )
...
This commit resolves the issue where `dt_boxes` was evaluated as a boolean
in Python 3.11, leading to a ValueError. The condition `if not dt_boxes:` is
replaced with `if dt_boxes.size == 0:` to explicitly check for an empty array,
ensuring compatibility with Python 3.11 and maintaining the intended functionality.
2023-11-22 20:10:32 +08:00
80459f59fe
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 ( #11280 )
...
使用gen_ocr_train_val_test.py分割recognition data後產生的train.txt、val.txt和test.txt每行label間多出一行空格，導致訓練時出現異常，移除換行\n後便可正常運行。
Co-authored-by: Wayne Huang <dinghsun@gmail.com>
2023-11-22 20:10:12 +08:00
68b384292b
v4 det cml configs ( #11258 )
...
* fixed several bugs
1. 修复了找不到模型的问题，原因是PPLCNetNew不在可选模型名称的列表内，将PPLCNetNew改为PPLCNetV3；
2. 修复了db_fpn.py解析in_channels的报错问题，db_fpn.py中按序列解析in_channels，而lcnetv3.py中det为false时输出是数值，为true才返回序列。在两个student模型的backbone中添加"det: true"；
3. 减少CPU占用，将cal_metric_during_train: true 改为 false；
4. 修复了训练过程中eval时的显存溢出问题。通过限制过大的测试数据可解决该问题，具体调整是，在eval→DetResizeForTest的配置中增加"limit_side_len: 960,limit_type: max"。
* 恢复cal_metric_during_train的设置
2023-11-17 20:38:25 +08:00
fb38d0be18
fix SAR inference, when batch size>1, norm_img_batch and valid_ratios shapes do not match ( #11238 )
2023-11-15 10:02:03 +08:00
73c08b7e4c
Add tipc for "ParseQ" method ( #10843 )
...
* Create rec_vit_parseq.yml
* Create train_infer_python.txt
2023-11-01 16:11:46 +08:00
bf6b100f85
update uygur dict ( #11125 )
2023-10-26 15:32:55 +08:00
987b3112cd
valid_out ( #11126 )
2023-10-25 15:35:10 +08:00
bf59c42a0f
fix satrn export for paddle2.5 ( #11096 )
2023-10-19 15:00:11 +08:00
a0218a8b83
[TIPC]update tipc scripts ( #11097 )
2023-10-18 20:05:52 +08:00
e3fc6393e0
[Cherry-pick] Cherry-pick from release/2.6 ( #11092 )
...
* Update recognition_en.md (#10059 )
ic15_dict.txt only have 36 digits
* Update ocr_rec.h (#9469 )
It is enough to include preprocess_op.h, we do not need to include ocr_cls.h.
* 补充num_classes注释说明 (#10073 )
ser_vi_layoutxlm_xfund_zh.yml中的Architecture.Backbone.num_classes所赋值会设置给Loss.num_classes，
由于采用BIO标注，假设字典中包含n个字段（包含other）时，则类别数为2n-1;假设字典中包含n个字段（不含other）时，则类别数为2n+1。
* Update algorithm_overview_en.md (#9747 )
Fix links to super-resolution algorithm docs
* 改进文档`deploy/hubserving/readme.md`和`doc/doc_ch/models_list.md` (#9110 )
* Update readme.md
* Update readme.md
* Update readme.md
* Update models_list.md
* trim trailling spaces @ `deploy/hubserving/readme_en.md`
* `s/shell/bash/` @ `deploy/hubserving/readme_en.md`
* Update `deploy/hubserving/readme_en.md` to sync with `deploy/hubserving/readme.md`
* Update deploy/hubserving/readme_en.md to sync with `deploy/hubserving/readme.md`
* Update deploy/hubserving/readme_en.md to sync with `deploy/hubserving/readme.md`
* Update `doc/doc_en/models_list_en.md` to sync with `doc/doc_ch/models_list_en.md`
* using Grammarly to weak `deploy/hubserving/readme_en.md`
* using Grammarly to tweak `doc/doc_en/models_list_en.md`
* `ocr_system` module will return with values of field `confidence`
* Update README_CN.md
* 修复测试服务中图片转Base64的引用地址错误。 (#8334 )
* Update application.md
* [Doc] Fix 404 link. (#10318 )
* Update PP-OCRv3_det_train.md
* Update knowledge_distillation.md
* Update config.md
* Fix fitz camelCase deprecation and .PDF not being recognized as pdf file (#10181 )
* Fix fitz camelCase deprecation and .PDF not being recognized as pdf file
* refactor get_image_file_list function
* Update customize.md (#10325 )
* Update FAQ.md (#10345 )
* Update FAQ.md (#10349 )
* Don't break overall processing on a bad image (#10216 )
* Add preprocessing common to OCR tasks (#10217 )
Add preprocessing to options
* [MLU] add mlu device for infer (#10249 )
* Create newfeature.md
* Update newfeature.md
* remove unused imported module, so can avoid PyInstaller packaged binary's start-time not found module error. (#10502 )
* CV套件建设专项活动 - 文字识别返回单字识别坐标 (#10515 )
* modification of return word box
* update_implements
* Update rec_postprocess.py
* Update utility.py
* Update README_ch.md
* revert README_ch.md update
* Fixed Layout recovery README file (#10493 )
Co-authored-by: Shubham Chambhare <shubhamchambhare@zoop.one>
* update_doc
* bugfix
---------
Co-authored-by: ChuongLoc <89434232+ChuongLoc@users.noreply.github.com>
Co-authored-by: Wang Xin <xinwang614@gmail.com>
Co-authored-by: tanjh <dtdhinjapan@gmail.com>
Co-authored-by: Louis Maddox <lmmx@users.noreply.github.com>
Co-authored-by: n0099 <n@n0099.net>
Co-authored-by: zhenliang li <37922155+shouyong@users.noreply.github.com>
Co-authored-by: itasli <ilyas.tasli@outlook.fr>
Co-authored-by: UserUnknownFactor <63057995+UserUnknownFactor@users.noreply.github.com>
Co-authored-by: PeiyuLau <135964669+PeiyuLau@users.noreply.github.com>
Co-authored-by: kerneltravel <kjpioo2006@gmail.com>
Co-authored-by: ToddBear <43341135+ToddBear@users.noreply.github.com>
Co-authored-by: Ligoml <39876205+Ligoml@users.noreply.github.com>
Co-authored-by: Shubham Chambhare <59397280+Shubham654@users.noreply.github.com>
Co-authored-by: Shubham Chambhare <shubhamchambhare@zoop.one>
Co-authored-by: andyj <87074272+andyjpaddle@users.noreply.github.com>
2023-10-18 17:37:23 +08:00
ce728358d7
Update algorithm_overview_en.md ( #10734 )
...
fix: tutorial link
2023-10-18 14:45:27 +08:00
60acb26abf
Update README.md ( #10733 )
...
typo error
2023-10-13 10:30:02 +08:00
c1134599e7
Update quickstart_en.md ( #10732 )
...
typo error
2023-10-13 10:28:34 +08:00
2213807323
refactored splitTrainVal and added multiOS path support ( #11069 )
2023-10-13 10:27:26 +08:00
d0d77fe273
Fixed a bug where changing "max_text_length" for VisionLAN did not change the length of the output text ( #11025 )
2023-10-12 11:07:07 +08:00
3786b27307
add cppd u14m train model and doc ( #11052 )
...
* add cppd u14m train model
* add cppd u14m train model and doc
2023-10-11 17:15:01 +08:00
f59d0929b0
Update how_to_do_kie_en.md ( #10731 )
...
fix: typo error
ID card is mentioned instead of ID No.
2023-10-11 10:08:42 +08:00
799c144ab3
update ppocrv4_framework ( #11048 )
2023-10-08 10:52:10 +08:00
6abf08a06e
fix import collection for py310 ( #11012 )
2023-09-27 15:40:16 +08:00
79dabcc9c7
Update kie_datasets_en.md ( #10735 )
...
fix: typo error correction
2023-09-27 11:14:33 +08:00
4ba32bc91c
support eval pre epoch ( #11003 )
2023-09-26 18:50:42 +08:00
e49e491417
add svtr large model ( #10937 )
...
* add svtr large model
* [WIP]add svtr large model
2023-09-26 14:38:29 +08:00
2751cb3a11
fix `cls_x` and `bbox_x` is possibly unbound ( #10991 )
2023-09-25 17:15:09 +08:00
8a52c99ad8
[New] add rec CPPD model ( #10990 )
...
* fix gris_sample data type bug when use fp16
* fix gris_sample data type bug when use fp16
* fix v4rec batchsize
* fix bug of hang when multi gpus training(sampler)
* add rec algorithm cppd
* delete cppd useless code
* update cppd bug
* add rec algorithm cppd
* update cppd trainedmodel url
* add cppd en doc
2023-09-25 15:43:45 +08:00
2fdac319b1
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中增加了对透明色的需求，由于在check_img中以及存在对单通道图像转三通道图像的处理，因此将该四通道图像处理成三通道的图像也放入该check_img，并统一三通道图像输出逻辑。 ( #10847 )
...
add:由于函数复杂度增高，增加注释。
2023-09-21 14:51:32 +08:00
ebc67db25b
rec_r45_abinet for export model ( #10892 )
...
* When exporting the inference model for ABINet, adapt to the 'image_shape' of ABINetRecResizeImg.
* restore h
2023-09-21 14:50:58 +08:00
5d68082518
rm fluid for paddle dev ( #10931 )
2023-09-19 10:16:35 +08:00
75d16610f4
Add new recognition method "ParseQ" ( #10836 )
...
* Update PP-OCRv4_introduction.md
* Update PP-OCRv4_introduction.md (#10616 )
* Update PP-OCRv4_introduction.md
* Update PP-OCRv4_introduction.md
* Update PP-OCRv4_introduction.md
* Update README.md
* Cherrypicking GH-10217 and GH-10216 to PaddlePaddle:Release/2.7 (#10655 )
* Don't break overall processing on a bad image
* Add preprocessing common to OCR tasks
Add preprocessing to options
* Update requirements.txt (#10656 )
added missing pyyaml library
* [TIPC]update xpu tipc script (#10658 )
* fix-typo (#10642 )
Co-authored-by: Dennis <dvorst@users.noreply.github.com>
Co-authored-by: shiyutang <34859558+shiyutang@users.noreply.github.com>
* 修改数据增强导致的DSR报错 (#10662 ) (#10681 )
* 修改数据增强导致的DSR报错
* 错误修改回滚
* Update algorithm_overview_en.md (#10670 )
Fixed simple spelling errors.
* Implement recoginition method ParseQ
* Document update for new recognition method ParseQ
* add prediction for parseq
* Update rec_vit_parseq.yml
* Update rec_r31_sar.yml
* Update rec_r31_sar.yml
* Update rec_r50_fpn_srn.yml
* Update rec_vit_parseq.py
* Update rec_vit_parseq.yml
* Update rec_parseq_head.py
* Update rec_img_aug.py
* Update rec_vit_parseq.yml
* Update __init__.py
* Update predict_rec.py
* Update paddleocr.py
* Update requirements.txt
* Update utility.py
* Update utility.py
---------
Co-authored-by: xiaoting <31891223+tink2123@users.noreply.github.com>
Co-authored-by: topduke <784990967@qq.com>
Co-authored-by: dyning <dyning.2003@163.com>
Co-authored-by: UserUnknownFactor <63057995+UserUnknownFactor@users.noreply.github.com>
Co-authored-by: itasli <ilyas.tasli@outlook.fr>
Co-authored-by: Kai Song <50285351+USTCKAY@users.noreply.github.com>
Co-authored-by: dvorst <87502756+dvorst@users.noreply.github.com>
Co-authored-by: Dennis <dvorst@users.noreply.github.com>
Co-authored-by: shiyutang <34859558+shiyutang@users.noreply.github.com>
Co-authored-by: Dec20B <1192152456@qq.com>
Co-authored-by: ncoffman <51147417+ncoffman@users.noreply.github.com>
2023-09-07 16:36:47 +08:00
ab86490138
Update algorithm_kie_vi_layoutxlm_en.md ( #10736 )
...
fix: typo error and created `2. Environment` header section which was missing before.
2023-09-05 15:42:09 +08:00
3a6fe48afd
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 image to the 3 channel BGR color image.)，因此修改该读取参数，以解决后续通道不匹配问题。 ( #10777 )
2023-08-31 14:31:28 +08:00
63f5787bfa
ch_PP-OCRv4_rec_distill.yml, fix KeyError: 'NRTRLabelDecode' ( #10761 )
2023-08-31 14:25:43 +08:00
e3cd343341
rec_r45_abinet.yml add max_length and image_size ( #10744 )
...
* rec_r45_abinet.yml add max_length and image_shape
* image_shape to image_size
2023-08-31 14:23:47 +08:00
66461c3325
Cherrypicking GH-10251 & GH-10181 to PaddleOCR:dygraph ( #10710 )
...
* fix undefined save_model_dir when using wandb (#10251 )
* fix undefined save_model_dir when using wand
* Update program.py
* Fix fitz camelCase deprecation and .PDF not being recognized as pdf file (#10181 )
* Fix fitz camelCase deprecation and .PDF not being recognized as pdf file
* refactor get_image_file_list function
2023-08-30 10:21:29 +08:00
1046c239cc
fix numpy speed ( #10773 )
2023-08-29 15:33:55 +08:00
b3912fcf7a
Cherrypicking GH-10217 and GH-10216 to PaddlePaddle:dygraph ( #10654 )
...
* Don't break overall processing on a bad image
* Add preprocessing common to OCR tasks
Add preprocessing to options
2023-08-21 16:33:03 +08:00
2bd552c801
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： ( #10628 )
...
* 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：
https://github.com/PaddlePaddle/PaddleOCR/issues/10533
* 删除了乱码字符
* 1.删除了乱码字符
* 1.删除了乱码字符
2023-08-21 10:25:47 +08:00
21da7f4c00
Update VQA to use the updated LayoutLM syntax from PaddleNLP ( #9791 )
...
* Update vqa_layoutlm.py
Update vqa to use the updated LayoutLM syntax from PaddleNLP
* Update vqa_layoutlm.py
2023-08-18 11:51:54 +08:00
dbf35bb714
onnxruntime support gpu ( #10668 )
...
* Update ch_PP-OCRv3_rec.yml
* Update ch_PP-OCRv3_rec_distillation.yml
* Update en_PP-OCRv3_rec.yml
* Update arabic_PP-OCRv3_rec.yml
* Update chinese_cht_PP-OCRv3_rec.yml
* Update cyrillic_PP-OCRv3_rec.yml
* Update devanagari_PP-OCRv3_rec.yml
* Update japan_PP-OCRv3_rec.yml
* Update ka_PP-OCRv3_rec.yml
* Update korean_PP-OCRv3_rec.yml
* Update latin_PP-OCRv3_rec.yml
* Update ta_PP-OCRv3_rec.yml
* Update te_PP-OCRv3_rec.yml
* Update utility.py
2023-08-17 18:26:20 +08:00
8f010ecf1e
修改数据增强导致的DSR报错 ( #10662 )
...
* 修改数据增强导致的DSR报错
* 错误修改回滚
2023-08-17 15:32:56 +08:00
549376ead5
[Cherry-pick] #10441 #10512 ( #10593 )
...
* fix memory leak (#10441 )
* fix memory leak
* update: Using smart pointers instead of raw pointers
* update: Usinig intuitive initialization of duration (#10512 )
2023-08-14 10:14:43 +08:00
56160986da
[bug fix]rm invalid params ( #10605 )
...
* add finetune en doc & test=document_fix
* fix dead link & test=document_fix
* fix dead link & test=document_fix
* update check img
* fix det res dtype
* update args default type & test=document_fix
* fix numpy version
* support numpy1.24.0
* fix doc & test=document_fix
* update doc
* update doc, test=document_fix
* fix pdf2word in whl, test=document_fix
* fix none res in recovery
* update version
* format code
* rm invalid params
2023-08-11 14:12:26 +08:00
2a4d551547
Fix seed passing issue of build_dataloader ( #10614 )
2023-08-11 14:11:13 +08:00
681467d4ea
[bug fix] fix none res in recovery ( #10603 )
...
* add finetune en doc & test=document_fix
* fix dead link & test=document_fix
* fix dead link & test=document_fix
* update check img
* fix det res dtype
* update args default type & test=document_fix
* fix numpy version
* support numpy1.24.0
* fix doc & test=document_fix
* update doc
* update doc, test=document_fix
* fix pdf2word in whl, test=document_fix
* fix none res in recovery
* update version
* format code
2023-08-10 16:55:26 +08:00
4a91a21245
compat_pillow ( #10596 )
2023-08-10 16:41:35 +08:00
TracebaK
gmm
xiaoting
RuohengMa
ficodex
黃鼎勲 (Ding Hsun Huang)
sylarwcy
Alexnet
ToddBear
fzhi
shiyutang
Kai Song
Sagar J
itasli
Viktor Bordiuzha
topduke
zhangyubo0722
Nyakku Shigure
Gmgge
xlg-go
wanghuancoder
UserUnknownFactor
WilliamQf
Sijun He
zhoujun
Dec20B
moehuster
andyj