From 82fceb5bf4bf41971c37346cd6ca2b4101478a21 Mon Sep 17 00:00:00 2001 From: cuicheng01 <45199522+cuicheng01@users.noreply.github.com> Date: Fri, 27 Nov 2020 17:55:11 +0800 Subject: [PATCH] Add GhostNet_x1_3_ssld_pretrained in master branch. (#420) Add GhostNet_x1_3_ssld_pretrained --- README.md | 4 +++- README_cn.md | 3 ++- docs/en/models/Mobile_en.md | 3 +++ docs/en/models/models_intro_en.md | 1 + docs/en/update_history_en.md | 4 ++++ docs/zh_CN/models/Mobile.md | 3 ++- docs/zh_CN/models/models_intro.md | 2 +- docs/zh_CN/update_history.md | 3 +++ 8 files changed, 19 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 31dbfbe26..4415c2c5d 100644 --- a/README.md +++ b/README.md @@ -7,7 +7,8 @@ PaddleClas is a toolset for image classification tasks prepared for the industry and academia. It helps users train better computer vision models and apply them in real scenarios. **Recent update** -- 2020.11.09 Add `InceptionV3` architecture and pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 79.1%. +- 2020.11.23 Add `GhostNet_x1_3_ssld` pretrained model, whose Top-1 Acc on ImageNet-1k dataset reaches 79.38%. +- 2020.11.09 Add `InceptionV3` architecture and pretrained model, whose Top-1 Acc on ImageNet-1k dataset reaches 79.1%. - 2020.10.12 Add Paddle-Lite demo. - 2020.10.10 Add cpp inference demo and improve FAQ tutorial. - 2020.09.17 Add `HRNet_W48_C_ssld` pretrained model, whose Top-1 Acc on ImageNet-1k dataset reaches 83.62%. Add `ResNet34_vd_ssld` pretrained model, whose Top-1 Acc on ImageNet-1k dataset reaches 79.72%. @@ -182,6 +183,7 @@ Accuracy and inference time metrics of Mobile series models are shown as follows | GhostNet_
x0_5 | 0.6688 | 0.8695 | 5.7143 | 0.082 | 2.6 | 10 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x0_5_pretrained.pdparams) | | GhostNet_
x1_0 | 0.7402 | 0.9165 | 13.5587 | 0.294 | 5.2 | 20 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_0_pretrained.pdparams) | | GhostNet_
x1_3 | 0.7579 | 0.9254 | 19.9825 | 0.44 | 7.3 | 29 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_pretrained.pdparams) | +| GhostNet_
x1_3_ssld | 0.7938 | 0.9449 | 19.9825 | 0.44 | 7.3 | 29 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_ssld_pretrained.tar) | diff --git a/README_cn.md b/README_cn.md index 8765593cf..cab541c26 100644 --- a/README_cn.md +++ b/README_cn.md @@ -7,6 +7,7 @@ 飞桨图像分类套件PaddleClas是飞桨为工业界和学术界所准备的一个图像分类任务的工具集，助力使用者训练出更好的视觉模型和应用落地。 **近期更新** +- 2020.11.23 添加`GhostNet_x1_3_ssld `模型，在ImageNet-1k上Top-1 Acc可达79.38%。 - 2020.11.18 添加图像分类[常见问题2020第一季第三期](./docs/zh_CN/faq_series/faq_2020_s1.md) 5个新问题，并且计划以后每周会更新，欢迎大家持续关注。 - 2020.11.09 添加`InceptionV3 `结构和模型，在ImageNet-1k上Top-1 Acc可达79.14%。 - 2020.10.12 添加Paddle-Lite demo。 @@ -187,7 +188,7 @@ ResNet及其Vd系列模型的精度、速度指标如下表所示，更多关于 | GhostNet_
x0_5 | 0.6688 | 0.8695 | 5.7143 | 0.082 | 2.6 | 10 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x0_5_pretrained.pdparams) | | GhostNet_
x1_0 | 0.7402 | 0.9165 | 13.5587 | 0.294 | 5.2 | 20 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_0_pretrained.pdparams) | | GhostNet_
x1_3 | 0.7579 | 0.9254 | 19.9825 | 0.44 | 7.3 | 29 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_pretrained.pdparams) | - +| GhostNet_
x1_3_ssld | 0.7938 | 0.9449 | 19.9825 | 0.44 | 7.3 | 29 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_ssld_pretrained.tar) | ### SEResNeXt与Res2Net系列 diff --git a/docs/en/models/Mobile_en.md b/docs/en/models/Mobile_en.md index 66b208541..fd4a2240e 100644 --- a/docs/en/models/Mobile_en.md +++ b/docs/en/models/Mobile_en.md @@ -62,6 +62,7 @@ Currently there are 32 pretrained models of the mobile series open source by Pad | GhostNet_x0_5 | 0.668 | 0.869 | 0.662 | 0.866 | 0.082 | 2.600 | | GhostNet_x1_0 | 0.740 | 0.916 | 0.739 | 0.914 | 0.294 | 5.200 | | GhostNet_x1_3 | 0.757 | 0.925 | 0.757 | 0.927 | 0.440 | 7.300 | +| GhostNet_x1_3_ssld | 0.794 | 0.945 | 0.757 | 0.927 | 0.440 | 7.300 | ## Inference speed and storage size based on SD855 @@ -104,6 +105,8 @@ Currently there are 32 pretrained models of the mobile series open source by Pad | GhostNet_x0_5 | 5.714 | 10.000 | | GhostNet_x1_0 | 13.558 | 20.000 | | GhostNet_x1_3 | 19.982 | 29.000 | +| GhostNet_x1_3_ssld | 19.982 | 29.000 | + ## Inference speed based on T4 GPU diff --git a/docs/en/models/models_intro_en.md b/docs/en/models/models_intro_en.md index 9782a8a63..60c4fbd78 100644 --- a/docs/en/models/models_intro_en.md +++ b/docs/en/models/models_intro_en.md @@ -98,6 +98,7 @@ python tools/infer/predict.py \ - [GhostNet_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x0_5_pretrained.pdparams) - [GhostNet_x1_0](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_0_pretrained.pdparams) - [GhostNet_x1_3](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_pretrained.pdparams) + - [GhostNet_x1_3_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_ssld_pretrained.tar) - SEResNeXt and Res2Net series diff --git a/docs/en/update_history_en.md b/docs/en/update_history_en.md index 4caa94937..011d12da5 100644 --- a/docs/en/update_history_en.md +++ b/docs/en/update_history_en.md @@ -1,5 +1,9 @@ # Release Notes + +- 2020.11.23 + * Add `GhostNet_x1_3_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 79.38%. + - 2020.11.09 * Add `InceptionV3` architecture and pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 79.1%. diff --git a/docs/zh_CN/models/Mobile.md b/docs/zh_CN/models/Mobile.md index 24bdf2392..708d7f3c8 100644 --- a/docs/zh_CN/models/Mobile.md +++ b/docs/zh_CN/models/Mobile.md @@ -63,7 +63,7 @@ GhostNet是华为于2020年提出的一种全新的轻量化网络结构，通 | GhostNet_x0_5 | 0.668 | 0.869 | 0.662 | 0.866 | 0.082 | 2.600 | | GhostNet_x1_0 | 0.740 | 0.916 | 0.739 | 0.914 | 0.294 | 5.200 | | GhostNet_x1_3 | 0.757 | 0.925 | 0.757 | 0.927 | 0.440 | 7.300 | - +| GhostNet_x1_3_ssld | 0.794 | 0.945 | 0.757 | 0.927 | 0.440 | 7.300 | ## 基于SD855的预测速度和存储大小 @@ -105,6 +105,7 @@ GhostNet是华为于2020年提出的一种全新的轻量化网络结构，通 | GhostNet_x0_5 | 5.714 | 10.000 | | GhostNet_x1_0 | 13.558 | 20.000 | | GhostNet_x1_3 | 19.982 | 29.000 | +| GhostNet_x1_3_ssld | 19.982 | 29.000 | ## 基于T4 GPU的预测速度 diff --git a/docs/zh_CN/models/models_intro.md b/docs/zh_CN/models/models_intro.md index 33c6b248b..451de9e9d 100644 --- a/docs/zh_CN/models/models_intro.md +++ b/docs/zh_CN/models/models_intro.md @@ -98,7 +98,7 @@ python tools/infer/predict.py \ - [GhostNet_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x0_5_pretrained.pdparams) - [GhostNet_x1_0](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_0_pretrained.pdparams) - [GhostNet_x1_3](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_pretrained.pdparams) - + - [GhostNet_x1_3_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/GhostNet_x1_3_ssld_pretrained.tar) - SEResNeXt与Res2Net系列 - ResNeXt系列[[7](#ref7)]([论文地址](https://arxiv.org/abs/1611.05431)) diff --git a/docs/zh_CN/update_history.md b/docs/zh_CN/update_history.md index 346838b89..5fbed8fb2 100644 --- a/docs/zh_CN/update_history.md +++ b/docs/zh_CN/update_history.md @@ -1,5 +1,8 @@ # 更新日志 +- 2020.11.23 + * 添加GhostNet_x1_3_ssld模型，在ImageNet上Top-1 Acc可达0.7938。 + - 2020.11.09 * 添加InceptionV3结构和模型，在ImageNet上Top-1 Acc可达0.791。