瀏覽代碼

[GCU] Update model list

EnflameGCU 11 月之前
父節點
當前提交
1f973f6645

+ 2 - 2
docs/other_devices_support/paddlepaddle_install_GCU.en.md

@@ -28,8 +28,8 @@ Download and install the wheel package released by PaddlePaddle within the docke
 * Download and install the wheel package.
 ```bash
 # Note: You need to install the CPU version of PaddlePaddle first
-python -m pip install paddlepaddle==3.0.0.dev20241127 -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/
-python -m pip install paddle_custom_gcu==3.0.0.dev20241127 -i https://www.paddlepaddle.org.cn/packages/nightly/gcu/
+python -m pip install paddlepaddle==3.0.0.dev20241202 -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/
+python -m pip install paddle_custom_gcu==3.0.0.dev20241203 -i https://www.paddlepaddle.org.cn/packages/nightly/gcu/
 ```
 * Verify the installation package. After installation, run the following command:
 ```bash

+ 2 - 2
docs/other_devices_support/paddlepaddle_install_GCU.md

@@ -28,8 +28,8 @@ bash TopsRider_i3x_*_deb_amd64.run --driver --no-auto-load
 * 下载并安装 wheel 包。
 ```bash
 # 注意需要先安装飞桨 cpu 版本
-python -m pip install paddlepaddle==3.0.0.dev20241127 -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/
-python -m pip install paddle_custom_gcu==3.0.0.dev20241127 -i https://www.paddlepaddle.org.cn/packages/nightly/gcu/
+python -m pip install paddlepaddle==3.0.0.dev20241202 -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/
+python -m pip install paddle_custom_gcu==3.0.0.dev20241203 -i https://www.paddlepaddle.org.cn/packages/nightly/gcu/
 ```
 * 验证安装包:安装完成之后,运行如下命令:
 ```bash

+ 107 - 0
docs/support_list/model_list_gcu.en.md

@@ -24,3 +24,110 @@ PaddleX incorporates multiple pipelines, each containing several modules, and ea
 </tbody>
 </table>
 <b>Note: The above accuracy metrics refer to Top-1 Accuracy on the [ImageNet-1k](https://www.image-net.org/index.php) validation set.</b>
+
+## Object Detection Module
+<table>
+<thead>
+<tr>
+<th>Model Name</th>
+<th>mAP (%)</th>
+<th>Model Size (M)</th>
+<th>Model Download Link</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-YOLOE_plus-L</td>
+<td>52.8</td>
+<td>185.3 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-L_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-M</td>
+<td>49.7</td>
+<td>83.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-M_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-M_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-S</td>
+<td>43.6</td>
+<td>28.3 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-S_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-X</td>
+<td>54.7</td>
+<td>349.4 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-X_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-X_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>RT-DETR-H</td>
+<td>56.3</td>
+<td>435.8 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-H_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>RT-DETR-L</td>
+<td>53.0</td>
+<td>113.7 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-L_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>RT-DETR-R18</td>
+<td>46.5</td>
+<td>70.7 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-R18_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-R18_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>RT-DETR-R50</td>
+<td>53.1</td>
+<td>149.1 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-R50_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-R50_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>RT-DETR-X</td>
+<td>54.8</td>
+<td>232.9 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-X_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-X_pretrained.pdparams">Trained Model</a></td></tr>
+</tbody>
+</table>
+<b>Note: The above accuracy metrics are for</b> [COCO2017](https://cocodataset.org/#home) <b>validation set mAP(0.5:0.95).</b>
+
+## Text Detection Module
+<table>
+<thead>
+<tr>
+<th>Model Name</th>
+<th>Detection Hmean (%)</th>
+<th>Model Size (M)</th>
+<th>Model Download Link</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-OCRv4_mobile_det</td>
+<td>77.79</td>
+<td>4.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_mobile_det_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_det_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>PP-OCRv4_server_det</td>
+<td>82.69</td>
+<td>100.1 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_server_det_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_det_pretrained.pdparams">Trained Model</a></td></tr>
+</tbody>
+</table>
+<b>Note: The above accuracy metrics are evaluated on PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, and handwritten scenarios, with 500 images for detection.</b>
+
+## Text Recognition Module
+<table>
+<thead>
+<tr>
+<th>Model Name</th>
+<th>Recognition Avg Accuracy (%)</th>
+<th>Model Size (M)</th>
+<th>Model Download Link</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-OCRv4_mobile_rec</td>
+<td>78.20</td>
+<td>10.6 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Trained Model</a></td></tr>
+<tr>
+<td>PP-OCRv4_server_rec</td>
+<td>79.20</td>
+<td>71.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td></tr>
+</tbody>
+</table>
+<b>Note: The above accuracy metrics are evaluated on PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, and handwritten scenarios, with 11,000 images for text recognition.</b>

+ 107 - 0
docs/support_list/model_list_gcu.md

@@ -24,3 +24,110 @@ PaddleX 内置了多条产线,每条产线都包含了若干模块,每个模
 </tbody>
 </table>
 <b>注:以上精度指标为</b>[ImageNet-1k](https://www.image-net.org/index.php)<b>验证集 Top1 Acc。</b>
+
+## 目标检测模块
+<table>
+<thead>
+<tr>
+<th>模型名称</th>
+<th>mAP(%)</th>
+<th>模型存储大小(M)</th>
+<th>模型下载链接</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-YOLOE_plus-L</td>
+<td>52.8</td>
+<td>185.3 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-L_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-M</td>
+<td>49.7</td>
+<td>83.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-M_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-M_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-S</td>
+<td>43.6</td>
+<td>28.3 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-S_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>PP-YOLOE_plus-X</td>
+<td>54.7</td>
+<td>349.4 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-X_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-X_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>RT-DETR-H</td>
+<td>56.3</td>
+<td>435.8 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-H_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>RT-DETR-L</td>
+<td>53.0</td>
+<td>113.7 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-L_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>RT-DETR-R18</td>
+<td>46.5</td>
+<td>70.7 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-R18_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-R18_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>RT-DETR-R50</td>
+<td>53.1</td>
+<td>149.1 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-R50_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-R50_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>RT-DETR-X</td>
+<td>54.8</td>
+<td>232.9 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-X_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-X_pretrained.pdparams">训练模型</a></td></tr>
+</tbody>
+</table>
+<b>注:以上精度指标为</b>[COCO2017](https://cocodataset.org/#home)<b>验证集 mAP(0.5:0.95)。</b>
+
+## 文本检测模块
+<table>
+<thead>
+<tr>
+<th>模型名称</th>
+<th>检测Hmean(%)</th>
+<th>模型存储大小(M)</th>
+<th>模型下载链接</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-OCRv4_mobile_det</td>
+<td>77.79</td>
+<td>4.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_mobile_det_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_det_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>PP-OCRv4_server_det</td>
+<td>82.69</td>
+<td>100.1M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_server_det_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_det_pretrained.pdparams">训练模型</a></td></tr>
+</tbody>
+</table>
+<b>注:以上精度指标的评估集是 PaddleOCR 自建的中文数据集,覆盖街景、网图、文档、手写多个场景,其中检测包含 500 张图片。</b>
+
+## 文本识别模块
+<table>
+<thead>
+<tr>
+<th>模型名称</th>
+<th>识别Avg Accuracy(%)</th>
+<th>模型存储大小(M)</th>
+<th>模型下载链接</th></tr>
+</thead>
+<tbody>
+<tr>
+<td>PP-OCRv4_mobile_rec</td>
+<td>78.20</td>
+<td>10.6 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td></tr>
+<tr>
+<td>PP-OCRv4_server_rec</td>
+<td>79.20</td>
+<td>71.2 M</td>
+<td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td></tr>
+</tbody>
+</table>
+<b>注:以上精度指标的评估集是 PaddleOCR 自建的中文数据集,覆盖街景、网图、文档、手写多个场景,其中文本识别包含 1.1w 张图片。</b>