mirrors
/
PaddleClas
mirror of https://github.com/PaddlePaddle/PaddleClas.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

docs: resize pic

pull/1570/head
gaotingquan 2021-12-16 09:21:36 +00:00
parent 53ec5f8e18
commit 45fc927f34
No known key found for this signature in database
GPG Key ID: F3EF7F42536A30B7
4 changed files with 38 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
docs/en/algorithm_introduction/ImageNet_models_en.md
View File

@ -8,17 +8,21 @@ 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/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png" width="800">
</div>
Curves of accuracy to the inference time and storage size of common mobile-side models are shown as follows.
Curves of accuracy to the inference time of common mobile-side models are shown as follows.
![](../../images/models/mobile_arm_storage.png)
![](../../images/models/mobile_arm_top1.png)
<div align="center">
<img src="../../images/models/mobile_arm_top1.png" width="800">
</div>
Curves of accuracy to the inference time of some VisionTransformer models are shown as follows.
![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png" width="800">
</div>
<a name="SSLD_pretrained_series"></a>
### SSLD pretrained models

12
docs/en/models/models_intro_en.md
View File

@ -23,11 +23,17 @@ python tools/infer/predict.py \
--batch_size=1
```
![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png" width="800">
</div>
![](../../images/models/mobile_arm_top1.png)
<div align="center">
<img src="../../images/models/mobile_arm_top1.png" width="800">
</div>
![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png" width="800">
</div>
> 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)

16
docs/zh_CN/algorithm_introduction/ImageNet_models.md
View File

@ -44,17 +44,21 @@
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/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png" width="800">
</div>
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。
常见移动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/mobile_arm_storage.png)
![](../../images/models/mobile_arm_top1.png)
<div align="center">
<img src="../../images/models/mobile_arm_top1.png" width="800">
</div>
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)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png" width="800">
</div>
<a name="2"></a>

13
docs/zh_CN/models/models_intro.md
View File

@ -20,12 +20,17 @@
* 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。
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png" width="800">
</div>
![](../../images/models/V100_benchmark/v100.fp32.bs1.main_fps_top1_s.png)
<div align="center">
<img src="../../images/models/mobile_arm_top1.png" width="800">
</div>
![](../../images/models/mobile_arm_top1.png)
![](../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png)
<div align="center">
<img src="../../images/models/V100_benchmark/v100.fp32.bs1.visiontransformer.png" width="800">
</div>
> 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)