Skip to content

Commit

Permalink
deploy: 99a90d1
Browse files Browse the repository at this point in the history
  • Loading branch information
SWHL committed Sep 18, 2023
1 parent 04341b9 commit b38ede3
Show file tree
Hide file tree
Showing 36 changed files with 3,350 additions and 1,564 deletions.
1,383 changes: 1,383 additions & 0 deletions docs/about_model/convert_model/index.html

Large diffs are not rendered by default.

114 changes: 76 additions & 38 deletions docs/about_model/index.html

Large diffs are not rendered by default.

13 changes: 11 additions & 2 deletions docs/about_model/index.xml
Original file line number Diff line number Diff line change
Expand Up @@ -21,12 +21,21 @@ pre_process: DetResizeForTest: limit_side_len: 736 limit_type: min NormalizeImag
</item>

<item>
<title>支持识别语言及自助转换模型</title>
<title>支持识别语言</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</link>
<pubDate>Mon, 11 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</guid>
<description>支持识别的语言 link 因为本项目依托于PaddleOCR,所以理论上PaddleOCR支持识别的模型,RapidOCR都是支持的。 中英文检测和识别(可以直接使用) link 因为中英文是最为常用的模型,所以在打包时,就默认将中英文识别的模型放到了rapidocr_onnxruntime和rapidocr_openvino中,直接pip安装即可使用。 其他语种检测和识别(需要转换) link PaddleOCR中已有文本检测模型列表:link PaddleOCR已有文本识别模型列表: link 除了slim和蒸馏过的模型,上面链接中的其他模型都可以转换为ONNX格式,通过RapidOCR快速部署。 自助转换使用教程 link ⚠️ 主要借助paddleocr_convert库来实现。 在线快速转换 link 通过Hugging Face上的应用,快速转换模型。整体界面是下面这个样子 离线安装库转换 link 安装paddleocr_convert pip install paddleocr_convert 命令行使用 用法: $ paddleocr_convert -h usage: paddleocr_convert [-h] [-p MODEL_PATH] [-o SAVE_DIR] [-txt_path TXT_PATH] optional arguments: -h, --help show this help message and exit -p MODEL_PATH, --model_path MODEL_PATH The inference model url or local path of paddleocr. e.g. https://paddleocr.bj.bcebos.com/PP- OCRv3/chinese/ch_PP-OCRv3_det_infer.tar or models/ch_PP-OCRv3_det_infer.tar -o SAVE_DIR, --save_dir SAVE_DIR The directory of saving the model.</description>
<description>简介 link 因为本项目依托于PaddleOCR,所以理论上PaddleOCR支持识别的模型,RapidOCR都是支持的。 中英文检测和识别(可以直接使用) link 因为中英文是最为常用的模型,所以在打包时,就默认将中英文识别的模型放到了rapidocr_onnxruntime和rapidocr_openvino中,直接pip安装即可使用。 其他语种检测和识别(需要转换) link PaddleOCR中已有文本检测模型列表:link PaddleOCR已有文本识别模型列表: link 除了slim和蒸馏过的模型,上面链接中的其他模型都可以转换为ONNX格式,通过RapidOCR快速部署。 </description>
</item>

<item>
<title>转换Paddle模型为ONNX</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/convert_model/</link>
<pubDate>Mon, 18 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/convert_model/</guid>
<description>简介 link Paddle格式模型的转换,主要借助paddle2onnx库实现。针对PaddleOCR中涉及到的相关模型,直接转换并不太方便。因此,推出了PaddleOCRModelConverter转换工具。 主要有两种使用方式: 方法一:在线转换,借助魔搭和Hugging Face两个平台,搭建在线转换demo; 方法二:离线转换,pip安装该工具,即可本地转换使用。 在线快速转换 link 魔搭 Hugging Face 离线转换 link</description>
</item>

</channel>
Expand Down
109 changes: 65 additions & 44 deletions docs/about_model/model_summary/index.html

Large diffs are not rendered by default.

436 changes: 75 additions & 361 deletions docs/about_model/support_language/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/changelog/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/changelog/rapidocr/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/changelog/rapidocr_api/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/changelog/rapidocr_web/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/faq/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/inference_summary/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/onnxruntime/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/onnxruntime/infer_optim/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/onnxruntime/onnxruntime-gpu/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/openvino/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/openvino/infer-gpu/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/inference_engine/openvino/infer/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/overview/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/packages/rapidocr/config_parameter/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/packages/rapidocr/optim_tutorial/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr/overview/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr/rapidocr_onnxruntime/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/packages/rapidocr/rapidocr_openvino/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr_api/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr_web/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/packages/rapidocr_web/rapidocr_web/index.html

Large diffs are not rendered by default.

103 changes: 62 additions & 41 deletions docs/packages/rapidocr_web/rapidocr_web_desktop/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/packages/rapidocr_web/rapidocr_web_nuitka/index.html

Large diffs are not rendered by default.

95 changes: 58 additions & 37 deletions docs/quickstart/index.html

Large diffs are not rendered by default.

99 changes: 60 additions & 39 deletions docs/related_projects/index.html

Large diffs are not rendered by default.

55 changes: 32 additions & 23 deletions index.xml
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
<description>Recent content on RapidOCR Documentation</description>
<generator>Hugo -- gohugo.io</generator>
<language>en-us</language>
<lastBuildDate>Wed, 13 Sep 2023 00:00:00 +0000</lastBuildDate><atom:link href="https://rapidai.github.io/RapidOCRDocs/index.xml" rel="self" type="application/rss+xml" />
<lastBuildDate>Mon, 18 Sep 2023 00:00:00 +0000</lastBuildDate><atom:link href="https://rapidai.github.io/RapidOCRDocs/index.xml" rel="self" type="application/rss+xml" />
<item>
<title>rapidocr</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/changelog/rapidocr/</link>
Expand Down Expand Up @@ -76,19 +76,6 @@ $ rapidocr_web -ip 0.0.0.0 -p 9003 浏览器打开 http://localhost:9003/,enjo
<description>🍜2023-05-22 api update: link 将API从ocrweb中解耦出来,作为单独模块维护,详情参见API rapidocr_web&amp;gt;0.1.6之后,将不支持pip install rapidocr_web[api]方式安装,可直接pip install rapidocr_api安装使用。 </description>
</item>

<item>
<title>不同版本模型之间比较</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</link>
<pubDate>Mon, 11 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</guid>
<description>各个版本ONNX模型效果对比(仅供参考) link⚠️注意: 以下测试结果均在自己构建测试集上评测所得,不代表在其他测试集上结果也是如此。
文本检测模型 link 评测采用的是TextDetMetric库 + 文本检测测试集,详情可以移步AI Studio运行查看。
⚠️注意:以下表格中推理时间是基于MacBook Pro M1运行所得,不同机器会有差别,请侧重查看彼此之间的比较。
指标计算都是在以下参数下计算得来,差别仅在于模型文件不同。
pre_process: DetResizeForTest: limit_side_len: 736 limit_type: min NormalizeImage: std: [0.229, 0.224, 0.225] mean: [0.485, 0.456, 0.406] scale: 1./255. order: hwc ToCHWImage: KeepKeys: keep_keys: [&amp;#39;image&amp;#39;, &amp;#39;shape&amp;#39;] post_process: thresh: 0.3 box_thresh: 0.5 max_candidates: 1000 unclip_ratio: 1.6 use_dilation: true score_mode: &amp;#34;fast&amp;#34; 模型 模型大小 Precision Recall H-mean Speed(s/img) ch_PP-OCRv4_det_infer.onnx 4.5M 0.6958 0.8608 0.7696 0.6176 ch_PP-OCRv4_det_server_infer.onnx 108M 0.7070 0.9330 0.8044 13.9348 ch_PP-OCRv3_det_infer.onnx 2.3M 0.</description>
</item>

<item>
<title>桌面版使用教程</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr_web/rapidocr_web_desktop/</link>
Expand All @@ -102,15 +89,6 @@ pre_process: DetResizeForTest: limit_side_len: 736 limit_type: min NormalizeImag
. ├── api-ms-win-core-console-l1-1-0.dll ├── api-ms-win-core-datetime-l1-1-0.dll ├── api-ms-win-core-debug-l1-1-0.dll ├── api-ms-win-core-errorhandling-l1-1-0.dll ├── api-ms-win-core-file-l1-1-0.dll ├── api-ms-win-core-file-l1-2-0.dll ├── api-ms-win-core-file-l2-1-0.dll ├── api-ms-win-core-handle-l1-1-0.dll ├── api-ms-win-core-heap-l1-1-0.dll ├── api-ms-win-core-interlocked-l1-1-0.dll ├── api-ms-win-core-libraryloader-l1-1-0.dll ├── api-ms-win-core-localization-l1-2-0.dll ├── api-ms-win-core-memory-l1-1-0.dll ├── api-ms-win-core-namedpipe-l1-1-0.dll ├── api-ms-win-core-processenvironment-l1-1-0.dll ├── api-ms-win-core-processthreads-l1-1-0.dll ├── api-ms-win-core-processthreads-l1-1-1.dll ├── api-ms-win-core-profile-l1-1-0.dll ├── api-ms-win-core-rtlsupport-l1-1-0.dll ├── api-ms-win-core-string-l1-1-0.dll ├── api-ms-win-core-synch-l1-1-0.dll ├── api-ms-win-core-synch-l1-2-0.dll ├── api-ms-win-core-sysinfo-l1-1-0.dll ├── api-ms-win-core-timezone-l1-1-0.dll ├── api-ms-win-core-util-l1-1-0.dll ├── api-ms-win-crt-conio-l1-1-0.dll ├── api-ms-win-crt-convert-l1-1-0.</description>
</item>

<item>
<title>支持识别语言及自助转换模型</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</link>
<pubDate>Mon, 11 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</guid>
<description>支持识别的语言 link 因为本项目依托于PaddleOCR,所以理论上PaddleOCR支持识别的模型,RapidOCR都是支持的。 中英文检测和识别(可以直接使用) link 因为中英文是最为常用的模型,所以在打包时,就默认将中英文识别的模型放到了rapidocr_onnxruntime和rapidocr_openvino中,直接pip安装即可使用。 其他语种检测和识别(需要转换) link PaddleOCR中已有文本检测模型列表:link PaddleOCR已有文本识别模型列表: link 除了slim和蒸馏过的模型,上面链接中的其他模型都可以转换为ONNX格式,通过RapidOCR快速部署。 自助转换使用教程 link ⚠️ 主要借助paddleocr_convert库来实现。 在线快速转换 link 通过Hugging Face上的应用,快速转换模型。整体界面是下面这个样子 离线安装库转换 link 安装paddleocr_convert pip install paddleocr_convert 命令行使用 用法: $ paddleocr_convert -h usage: paddleocr_convert [-h] [-p MODEL_PATH] [-o SAVE_DIR] [-txt_path TXT_PATH] optional arguments: -h, --help show this help message and exit -p MODEL_PATH, --model_path MODEL_PATH The inference model url or local path of paddleocr. e.g. https://paddleocr.bj.bcebos.com/PP- OCRv3/chinese/ch_PP-OCRv3_det_infer.tar or models/ch_PP-OCRv3_det_infer.tar -o SAVE_DIR, --save_dir SAVE_DIR The directory of saving the model.</description>
</item>

<item>
<title>Nuitka打包rapidocr_web操作文档</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr_web/rapidocr_web_nuitka/</link>
Expand Down Expand Up @@ -198,6 +176,19 @@ from pathlib import Path import cv2 from rapidocr_onnxruntime import RapidOCR #
有值:([[文本框坐标], 文本内容, 置信度], 推理时间),e.g. [[左上, 右上, 右下, 左下], &#39;小明&#39;, &#39;0.</description>
</item>

<item>
<title>不同版本模型之间比较</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</link>
<pubDate>Mon, 11 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</guid>
<description>各个版本ONNX模型效果对比(仅供参考) link⚠️注意: 以下测试结果均在自己构建测试集上评测所得,不代表在其他测试集上结果也是如此。
文本检测模型 link 评测采用的是TextDetMetric库 + 文本检测测试集,详情可以移步AI Studio运行查看。
⚠️注意:以下表格中推理时间是基于MacBook Pro M1运行所得,不同机器会有差别,请侧重查看彼此之间的比较。
指标计算都是在以下参数下计算得来,差别仅在于模型文件不同。
pre_process: DetResizeForTest: limit_side_len: 736 limit_type: min NormalizeImage: std: [0.229, 0.224, 0.225] mean: [0.485, 0.456, 0.406] scale: 1./255. order: hwc ToCHWImage: KeepKeys: keep_keys: [&amp;#39;image&amp;#39;, &amp;#39;shape&amp;#39;] post_process: thresh: 0.3 box_thresh: 0.5 max_candidates: 1000 unclip_ratio: 1.6 use_dilation: true score_mode: &amp;#34;fast&amp;#34; 模型 模型大小 Precision Recall H-mean Speed(s/img) ch_PP-OCRv4_det_infer.onnx 4.5M 0.6958 0.8608 0.7696 0.6176 ch_PP-OCRv4_det_server_infer.onnx 108M 0.7070 0.9330 0.8044 13.9348 ch_PP-OCRv3_det_infer.onnx 2.3M 0.</description>
</item>

<item>
<title>调优尝试教程</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr/optim_tutorial/</link>
Expand All @@ -219,6 +210,24 @@ from pathlib import Path import cv2 from rapidocr_onnxruntime import RapidOCR #
Det部分 link 参数名称 取值范围 默认值 作用 use_cuda bool false 是否使用CUDA,加速推理 limit_side_len - 736 限制图像边的长度的像素值 limit_type [min, max] min 限制图像的最小边长度还是最大边为limit_side_len 示例解释:当limit_type=min和limit_side_len=736时,图像最小边小于736时,会将图像最小边拉伸到736,另一边则按图像原始比例等比缩放。 thresh [0, 1] 0.3 图像中文字部分和背景部分分割阈值值越大,文字部分会越小 box_thresh [0, 1] 0.5 文本检测所得框是否保留的阈值,值越大,召回率越低 max_candidates - 1000 图像中最大可检测到的文本框数目,一般够用 unclip_ratio [1.6, 2.0] 1.6 控制文本检测框的大小,值越大,检测框整体越大 use_dilation bool true 是否使用形态学中的膨胀操作,一般采用默认值即可 score_mode string fast fast是求rectangle区域的平均分数,容易造成弯曲文本漏检,slow是求polygon区域的平均分数,会更准确,但速度有所降低,可按需选择 Cls部分 link 参数名称 取值范围 默认值 作用 cls_img_shape - [3, 48, 192] 输入方向分类模型的图像Shape(CHW) cls_batch_num - 6 批次推理的batch大小,一般采用默认值即可,太大并没有明显提速,效果还可能会差 cls_thresh [0, 1] 0.</description>
</item>

<item>
<title>支持识别语言</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</link>
<pubDate>Mon, 11 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</guid>
<description>简介 link 因为本项目依托于PaddleOCR,所以理论上PaddleOCR支持识别的模型,RapidOCR都是支持的。 中英文检测和识别(可以直接使用) link 因为中英文是最为常用的模型,所以在打包时,就默认将中英文识别的模型放到了rapidocr_onnxruntime和rapidocr_openvino中,直接pip安装即可使用。 其他语种检测和识别(需要转换) link PaddleOCR中已有文本检测模型列表:link PaddleOCR已有文本识别模型列表: link 除了slim和蒸馏过的模型,上面链接中的其他模型都可以转换为ONNX格式,通过RapidOCR快速部署。 </description>
</item>

<item>
<title>转换Paddle模型为ONNX</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/about_model/convert_model/</link>
<pubDate>Mon, 18 Sep 2023 00:00:00 +0000</pubDate>

<guid>https://rapidai.github.io/RapidOCRDocs/docs/about_model/convert_model/</guid>
<description>简介 link Paddle格式模型的转换,主要借助paddle2onnx库实现。针对PaddleOCR中涉及到的相关模型,直接转换并不太方便。因此,推出了PaddleOCRModelConverter转换工具。 主要有两种使用方式: 方法一:在线转换,借助魔搭和Hugging Face两个平台,搭建在线转换demo; 方法二:离线转换,pip安装该工具,即可本地转换使用。 在线快速转换 link 魔搭 Hugging Face 离线转换 link</description>
</item>

<item>
<title>相关项目</title>
<link>https://rapidai.github.io/RapidOCRDocs/docs/related_projects/</link>
Expand Down
17 changes: 10 additions & 7 deletions sitemap.xml
Original file line number Diff line number Diff line change
Expand Up @@ -28,18 +28,12 @@
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/changelog/rapidocr_api/</loc>
<lastmod>2023-09-13T10:00:48+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</loc>
<lastmod>2023-09-16T07:58:31+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr_web/rapidocr_web_desktop/</loc>
<lastmod>2023-09-18T09:31:02+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/inference_engine/onnxruntime/</loc>
<lastmod>2023-09-14T22:11:20+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</loc>
<lastmod>2023-09-18T08:45:13+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr_web/rapidocr_web_nuitka/</loc>
<lastmod>2023-09-18T08:45:13+08:00</lastmod>
Expand Down Expand Up @@ -70,6 +64,9 @@
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/packages/</loc>
<lastmod>2023-09-15T11:28:10+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/model_summary/</loc>
<lastmod>2023-09-18T13:50:47+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr/optim_tutorial/</loc>
<lastmod>2023-09-18T08:45:13+08:00</lastmod>
Expand All @@ -79,9 +76,15 @@
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/packages/rapidocr/config_parameter/</loc>
<lastmod>2023-09-18T09:31:02+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/support_language/</loc>
<lastmod>2023-09-18T13:50:47+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/</loc>
<lastmod>2023-09-15T11:28:10+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/about_model/convert_model/</loc>
<lastmod>2023-09-18T13:50:47+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/docs/changelog/</loc>
<lastmod>2023-09-15T11:28:10+08:00</lastmod>
Expand All @@ -99,7 +102,7 @@
<lastmod>2023-09-18T08:45:13+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/</loc>
<lastmod>2023-09-18T09:31:02+08:00</lastmod>
<lastmod>2023-09-18T13:50:47+08:00</lastmod>
</url><url>
<loc>https://rapidai.github.io/RapidOCRDocs/categories/</loc>
</url><url>
Expand Down

0 comments on commit b38ede3

Please sign in to comment.