diff --git a/docs/en/algorithm_introduction/ImageNet_models_en.md b/docs/en/algorithm_introduction/ImageNet_models_en.md index 9a2f0b955..a3471c5f3 100644 --- a/docs/en/algorithm_introduction/ImageNet_models_en.md +++ b/docs/en/algorithm_introduction/ImageNet_models_en.md @@ -8,14 +8,17 @@ Based on the ImageNet-1k classification dataset, the 35 classification network s Curves of accuracy to the inference time of common server-side models are shown as follows. -![](../images/models/T4_benchmark/t4.fp32.bs1.main_fps_top1.png) - +![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png) Curves of accuracy to the inference time and storage size of common mobile-side models are shown as follows. -![](../images/models/mobile_arm_storage.png) +![](../../images/models/mobile_arm_storage.png) -![](../images/models/mobile_arm_top1.png) +![](../../images/models/mobile_arm_top1.png) + +Curves of accuracy to the inference time of some VisionTransformer models are shown as follows. + +![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png) ### SSLD pretrained models diff --git a/docs/en/models/models_intro_en.md b/docs/en/models/models_intro_en.md index 677d742d6..f932adaee 100644 --- a/docs/en/models/models_intro_en.md +++ b/docs/en/models/models_intro_en.md @@ -5,7 +5,7 @@ Based on the ImageNet1k classification dataset, the 29 classification network structures supported by PaddleClas and the corresponding 134 image classification pretrained models are shown below. Training trick, a brief introduction to each series of network structures, and performance evaluation will be shown in the corresponding chapters. ## Evaluation environment -* CPU evaluation environment is based on Snapdragon 855 (SD855). +* Arm evaluation environment is based on Snapdragon 855 (SD855). * The GPU evaluation environment is based on V100 and TensorRT, and the evaluation script is as follows. ```shell @@ -23,12 +23,11 @@ python tools/infer/predict.py \ --batch_size=1 ``` -![](../../images/models/T4_benchmark/t4.fp32.bs4.main_fps_top1.png) - -![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.jpg) +![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png) ![](../../images/models/mobile_arm_top1.png) +![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png) > If you think this document is helpful to you, welcome to give a star to our project:[https://github.com/PaddlePaddle/PaddleClas](https://github.com/PaddlePaddle/PaddleClas) diff --git a/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.jpg b/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.jpg deleted file mode 100644 index 44a101c67..000000000 Binary files a/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.jpg and /dev/null differ diff --git a/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png b/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png new file mode 100644 index 000000000..ab7f3eb37 Binary files /dev/null and b/docs/images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png differ diff --git a/docs/images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png b/docs/images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png new file mode 100644 index 000000000..500961912 Binary files /dev/null and b/docs/images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png differ diff --git a/docs/images/models/mobile_arm_top1.png b/docs/images/models/mobile_arm_top1.png index 87afa98ef..9156764dc 100644 Binary files a/docs/images/models/mobile_arm_top1.png and b/docs/images/models/mobile_arm_top1.png differ diff --git a/docs/zh_CN/algorithm_introduction/ImageNet_models.md b/docs/zh_CN/algorithm_introduction/ImageNet_models.md index 6d05078c9..3b6d7b4ba 100644 --- a/docs/zh_CN/algorithm_introduction/ImageNet_models.md +++ b/docs/zh_CN/algorithm_introduction/ImageNet_models.md @@ -44,8 +44,7 @@ 常见服务器端模型的精度指标与其预测耗时的变化曲线如下图所示。 -![](../../images/models/T4_benchmark/t4.fp32.bs1.main_fps_top1.png) - +![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png) 常见移动端模型的精度指标与其预测耗时、模型存储大小的变化曲线如下图所示。 @@ -53,6 +52,10 @@ ![](../../images/models/mobile_arm_top1.png) +部分VisionTransformer模型的精度指标与其预测耗时的变化曲线如下图所示。 + +![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png) + ## 2. SSLD 知识蒸馏预训练模型 diff --git a/docs/zh_CN/models/models_intro.md b/docs/zh_CN/models/models_intro.md index 0c30be7e4..b141a8af4 100644 --- a/docs/zh_CN/models/models_intro.md +++ b/docs/zh_CN/models/models_intro.md @@ -16,17 +16,16 @@ ## 2. 评估环境 -* CPU 的评估环境基于骁龙 855（SD855）。 +* Arm 的评估环境基于骁龙 855（SD855）。 * Intel CPU 的评估环境基于 Intel(R) Xeon(R) Gold 6148。 * GPU 评估环境基于 V100 和 TensorRT。 -![](../../images/models/T4_benchmark/t4.fp32.bs4.main_fps_top1.png) - -![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.jpg) +![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png) ![](../../images/models/mobile_arm_top1.png) +![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png) > 如果您觉得此文档对您有帮助，欢迎 star 我们的项目：[https://github.com/PaddlePaddle/PaddleClas](https://github.com/PaddlePaddle/PaddleClas)