Skip to content

P
PaddleClas

PaddlePaddle / PaddleClas
大约 1 年 前同步成功

通知 115
Star 4999
Fork 1114
P
PaddleClas
未验证 提交 f9413d1a 编写于 作者: L littletomatodonkey 提交者： GitHub
浏览文件
操作

Merge pull request #278 from littletomatodonkey/static/fix_readme_en 

fix readme en
上级 1adf8f4e 01f5042e
展开全部 显示空白变更内容
内联 并排
Showing with 314 addition and 127 deletion
README.md
浏览文件 @ f9413d1a
...@@ -7,6 +7,7 @@ ...@@ -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。 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** **近期更新**
- 2020.09.17 添加HRNet_W48_C_ssld模型，在ImageNet上Top-1 Acc可达0.836；添加ResNet34_vd_ssld模型，在ImageNet上Top-1 Acc可达0.797。
- 2020.09.07 添加HRNet_W18_C_ssld模型，在ImageNet上Top-1 Acc可达0.81162；添加MobileNetV3_small_x0_35_ssld模型，在ImageNet上Top-1 Acc可达0.5555。 - 2020.09.07 添加HRNet_W18_C_ssld模型，在ImageNet上Top-1 Acc可达0.81162；添加MobileNetV3_small_x0_35_ssld模型，在ImageNet上Top-1 Acc可达0.5555。
- 2020.07.14 添加Res2Net200_vd_26w_4s_ssld模型，在ImageNet上Top-1 Acc可达85.13%；添加Fix_ResNet50_vd_ssld_v2模型，在ImageNet上Top-1 Acc可达84.0%。 - 2020.07.14 添加Res2Net200_vd_26w_4s_ssld模型，在ImageNet上Top-1 Acc可达85.13%；添加Fix_ResNet50_vd_ssld_v2模型，在ImageNet上Top-1 Acc可达84.0%。
- 2020.06.17 添加英文文档。 - 2020.06.17 添加英文文档。
...@@ -17,7 +18,7 @@ ...@@ -17,7 +18,7 @@
## 特性 ## 特性
- 丰富的模型库：基于ImageNet1k分类数据集，PaddleClas提供了24个系列的分类网络结构和训练配置，121个预训练模型和性能评估。 - 丰富的模型库：基于ImageNet1k分类数据集，PaddleClas提供了24个系列的分类网络结构和训练配置，122个预训练模型和性能评估。
- SSLD知识蒸馏：基于该方案蒸馏模型的识别准确率普遍提升3%以上。 - SSLD知识蒸馏：基于该方案蒸馏模型的识别准确率普遍提升3%以上。
...@@ -41,8 +42,9 @@ ...@@ -41,8 +42,9 @@
- [ResNet及其Vd系列](#ResNet及其Vd系列) - [ResNet及其Vd系列](#ResNet及其Vd系列)
- [移动端系列](#移动端系列) - [移动端系列](#移动端系列)
- [SEResNeXt与Res2Net系列](#SEResNeXt与Res2Net系列) - [SEResNeXt与Res2Net系列](#SEResNeXt与Res2Net系列)
- [Inception系列](#Inception系列)
- [DPN与DenseNet系列](#DPN与DenseNet系列) - [DPN与DenseNet系列](#DPN与DenseNet系列)
- [HRNet](HRNet系列)
- [Inception系列](#Inception系列)
- [EfficientNet与ResNeXt101_wsl系列](#EfficientNet与ResNeXt101_wsl系列) - [EfficientNet与ResNeXt101_wsl系列](#EfficientNet与ResNeXt101_wsl系列)
- [ResNeSt与RegNet系列](#ResNeSt与RegNet系列) - [ResNeSt与RegNet系列](#ResNeSt与RegNet系列)
- 模型训练/评估 - 模型训练/评估
...@@ -103,6 +105,7 @@ ResNet及其Vd系列模型的精度、速度指标如下表所示，更多关于 ...@@ -103,6 +105,7 @@ ResNet及其Vd系列模型的精度、速度指标如下表所示，更多关于
| ResNet18_vd | 0.7226 | 0.9080 | 1.54557 | 3.85363 | 4.14 | 11.71 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar) | | ResNet18_vd | 0.7226 | 0.9080 | 1.54557 | 3.85363 | 4.14 | 11.71 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar) |
| ResNet34 | 0.7457 | 0.9214 | 2.34957 | 5.89821 | 7.36 | 21.8 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_pretrained.tar) | | ResNet34 | 0.7457 | 0.9214 | 2.34957 | 5.89821 | 7.36 | 21.8 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_pretrained.tar) |
| ResNet34_vd | 0.7598 | 0.9298 | 2.43427 | 6.22257 | 7.39 | 21.82 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar) | | ResNet34_vd | 0.7598 | 0.9298 | 2.43427 | 6.22257 | 7.39 | 21.82 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar) |
| ResNet34_vd_ssld | 0.7972 | 0.9490 | 2.43427 | 6.22257 | 7.39 | 21.82 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_ssld_pretrained.tar) |
| ResNet50 | 0.7650 | 0.9300 | 3.47712 | 7.84421 | 8.19 | 25.56 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_pretrained.tar) | | ResNet50 | 0.7650 | 0.9300 | 3.47712 | 7.84421 | 8.19 | 25.56 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_pretrained.tar) |
| ResNet50_vc | 0.7835 | 0.9403 | 3.52346 | 8.10725 | 8.67 | 25.58 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar) | | ResNet50_vc | 0.7835 | 0.9403 | 3.52346 | 8.10725 | 8.67 | 25.58 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar) |
| ResNet50_vd | 0.7912 | 0.9444 | 3.53131 | 8.09057 | 8.67 | 25.58 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar) | | ResNet50_vd | 0.7912 | 0.9444 | 3.53131 | 8.09057 | 8.67 | 25.58 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar) |
...@@ -235,6 +238,7 @@ HRNet系列模型的精度、速度指标如下表所示，更多关于该系列 ...@@ -235,6 +238,7 @@ HRNet系列模型的精度、速度指标如下表所示，更多关于该系列
| HRNet_W40_C | 0.7877 | 0.9447 | 12.12202 | 25.68184 | 25.41 | 57.55 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar) | | HRNet_W40_C | 0.7877 | 0.9447 | 12.12202 | 25.68184 | 25.41 | 57.55 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar) |
| HRNet_W44_C | 0.7900 | 0.9451 | 13.19858 | 32.25202 | 29.79 | 67.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar) | | HRNet_W44_C | 0.7900 | 0.9451 | 13.19858 | 32.25202 | 29.79 | 67.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar) |
| HRNet_W48_C | 0.7895 | 0.9442 | 13.70761 | 34.43572 | 34.58 | 77.47 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar) | | HRNet_W48_C | 0.7895 | 0.9442 | 13.70761 | 34.43572 | 34.58 | 77.47 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar) |
| HRNet_W48_C_ssld | 0.8363 | 0.9682 | 13.70761 | 34.43572 | 34.58 | 77.47 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar) |
| HRNet_W64_C | 0.7930 | 0.9461 | 17.57527 | 47.9533 | 57.83 | 128.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar) | | HRNet_W64_C | 0.7930 | 0.9461 | 17.57527 | 47.9533 | 57.83 | 128.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar) |
...@@ -257,7 +261,7 @@ Inception系列模型的精度、速度指标如下表所示，更多关于该 ...@@ -257,7 +261,7 @@ Inception系列模型的精度、速度指标如下表所示，更多关于该
<a name="EfficientNet与ResNeXt101_wsl系列"></a> <a name="EfficientNet与ResNeXt101_wsl系列"></a>
### EfficientNet与ResNeXt101_wsl系列 ### EfficientNet与ResNeXt101_wsl系列
EfficientNet与ResNeXt101_wsl系列模型的精度、速度指标如下表所示，更多关于该系列的模型介绍可以参考：[EfficientNet与ResNeXt101_wsl系列模型文档](./docs/zh_CN/models/Inception.md)。 EfficientNet与ResNeXt101_wsl系列模型的精度、速度指标如下表所示，更多关于该系列的模型介绍可以参考：[EfficientNet与ResNeXt101_wsl系列模型文档](./docs/zh_CN/models/EfficientNet_and_ResNeXt101_wsl.md)。
| 模型 | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | Flops(G) | Params(M) | 下载地址 | | 模型 | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | Flops(G) | Params(M) | 下载地址 |
......
README_en.md
浏览文件 @ f9413d1a
此差异已折叠。
docs/en/models/HRNet_en.md
浏览文件 @ f9413d1a
...@@ -28,6 +28,7 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC ...@@ -28,6 +28,7 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC
| HRNet_W40_C | 0.788 | 0.945 | 0.789 | 0.945 | 25.410 | 57.550 | | HRNet_W40_C | 0.788 | 0.945 | 0.789 | 0.945 | 25.410 | 57.550 |
| HRNet_W44_C | 0.790 | 0.945 | 0.789 | 0.944 | 29.790 | 67.060 | | HRNet_W44_C | 0.790 | 0.945 | 0.789 | 0.944 | 29.790 | 67.060 |
| HRNet_W48_C | 0.790 | 0.944 | 0.793 | 0.945 | 34.580 | 77.470 | | HRNet_W48_C | 0.790 | 0.944 | 0.793 | 0.945 | 34.580 | 77.470 |
| HRNet_W48_C_ssld | 0.836 | 0.968 | 0.793 | 0.945 | 34.580 | 77.470 |
| HRNet_W64_C | 0.793 | 0.946 | 0.795 | 0.946 | 57.830 | 128.060 | | HRNet_W64_C | 0.793 | 0.946 | 0.795 | 0.946 | 57.830 | 128.060 |
...@@ -42,6 +43,7 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC ...@@ -42,6 +43,7 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC
| HRNet_W40_C | 224 | 256 | 10.739 | | HRNet_W40_C | 224 | 256 | 10.739 |
| HRNet_W44_C | 224 | 256 | 11.497 | | HRNet_W44_C | 224 | 256 | 11.497 |
| HRNet_W48_C | 224 | 256 | 12.165 | | HRNet_W48_C | 224 | 256 | 12.165 |
| HRNet_W48_C_ssld | 224 | 256 | 12.165 |
| HRNet_W64_C | 224 | 256 | 15.003 | | HRNet_W64_C | 224 | 256 | 15.003 |
...@@ -58,4 +60,5 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC ...@@ -58,4 +60,5 @@ At present, there are 7 pretrained models of such models open-sourced by PaddleC
| HRNet_W40_C | 224 | 256 | 11.4229 | 19.1595 | 30.47984 | 12.12202 | 25.68184 | 48.90623 | | HRNet_W40_C | 224 | 256 | 11.4229 | 19.1595 | 30.47984 | 12.12202 | 25.68184 | 48.90623 |
| HRNet_W44_C | 224 | 256 | 12.25778 | 22.75456 | 32.61275 | 13.19858 | 32.25202 | 59.09871 | | HRNet_W44_C | 224 | 256 | 12.25778 | 22.75456 | 32.61275 | 13.19858 | 32.25202 | 59.09871 |
| HRNet_W48_C | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 | | HRNet_W48_C | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 |
| HRNet_W48_C_ssld | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 |
| HRNet_W64_C | 224 | 256 | 15.10428 | 27.68901 | 40.4198 | 17.57527 | 47.9533 | 97.11228 | | HRNet_W64_C | 224 | 256 | 15.10428 | 27.68901 | 40.4198 | 17.57527 | 47.9533 | 97.11228 |
docs/en/models/ResNet_and_vd_en.md
浏览文件 @ f9413d1a
...@@ -32,6 +32,7 @@ As can be seen from the above curves, the higher the number of layers, the highe ...@@ -32,6 +32,7 @@ As can be seen from the above curves, the higher the number of layers, the highe
| ResNet18_vd | 0.723 | 0.908 | | | 4.140 | 11.710 | | ResNet18_vd | 0.723 | 0.908 | | | 4.140 | 11.710 |
| ResNet34 | 0.746 | 0.921 | 0.732 | 0.913 | 7.360 | 21.800 | | ResNet34 | 0.746 | 0.921 | 0.732 | 0.913 | 7.360 | 21.800 |
| ResNet34_vd | 0.760 | 0.930 | | | 7.390 | 21.820 | | ResNet34_vd | 0.760 | 0.930 | | | 7.390 | 21.820 |
| ResNet34_vd_ssld | 0.797 | 0.949 | | | 7.390 | 21.820 |
| ResNet50 | 0.765 | 0.930 | 0.760 | 0.930 | 8.190 | 25.560 | | ResNet50 | 0.765 | 0.930 | 0.760 | 0.930 | 8.190 | 25.560 |
| ResNet50_vc | 0.784 | 0.940 | | | 8.670 | 25.580 | | ResNet50_vc | 0.784 | 0.940 | | | 8.670 | 25.580 |
| ResNet50_vd | 0.791 | 0.944 | 0.792 | 0.946 | 8.670 | 25.580 | | ResNet50_vd | 0.791 | 0.944 | 0.792 | 0.946 | 8.670 | 25.580 |
...@@ -57,6 +58,7 @@ As can be seen from the above curves, the higher the number of layers, the highe ...@@ -57,6 +58,7 @@ As can be seen from the above curves, the higher the number of layers, the highe
| ResNet18_vd | 224 | 256 | 1.603 | | ResNet18_vd | 224 | 256 | 1.603 |
| ResNet34 | 224 | 256 | 2.272 | | ResNet34 | 224 | 256 | 2.272 |
| ResNet34_vd | 224 | 256 | 2.343 | | ResNet34_vd | 224 | 256 | 2.343 |
| ResNet34_vd_ssld | 224 | 256 | 2.343 |
| ResNet50 | 224 | 256 | 2.939 | | ResNet50 | 224 | 256 | 2.939 |
| ResNet50_vc | 224 | 256 | 3.041 | | ResNet50_vc | 224 | 256 | 3.041 |
| ResNet50_vd | 224 | 256 | 3.165 | | ResNet50_vd | 224 | 256 | 3.165 |
...@@ -78,6 +80,7 @@ As can be seen from the above curves, the higher the number of layers, the highe ...@@ -78,6 +80,7 @@ As can be seen from the above curves, the higher the number of layers, the highe
| ResNet18_vd | 224 | 256 | 1.39593 | 2.69063 | 3.88267 | 1.54557 | 3.85363 | 6.88121 | | ResNet18_vd | 224 | 256 | 1.39593 | 2.69063 | 3.88267 | 1.54557 | 3.85363 | 6.88121 |
| ResNet34 | 224 | 256 | 2.23092 | 4.10205 | 5.54904 | 2.34957 | 5.89821 | 10.73451 | | ResNet34 | 224 | 256 | 2.23092 | 4.10205 | 5.54904 | 2.34957 | 5.89821 | 10.73451 |
| ResNet34_vd | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 | | ResNet34_vd | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 |
| ResNet34_vd | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 |
| ResNet50 | 224 | 256 | 2.63824 | 4.63802 | 7.02444 | 3.47712 | 7.84421 | 13.90633 | | ResNet50 | 224 | 256 | 2.63824 | 4.63802 | 7.02444 | 3.47712 | 7.84421 | 13.90633 |
| ResNet50_vc | 224 | 256 | 2.67064 | 4.72372 | 7.17204 | 3.52346 | 8.10725 | 14.45577 | | ResNet50_vc | 224 | 256 | 2.67064 | 4.72372 | 7.17204 | 3.52346 | 8.10725 | 14.45577 |
| ResNet50_vd | 224 | 256 | 2.65164 | 4.84109 | 7.46225 | 3.53131 | 8.09057 | 14.45965 | | ResNet50_vd | 224 | 256 | 2.65164 | 4.84109 | 7.46225 | 3.53131 | 8.09057 | 14.45965 |
......
docs/en/models/models_intro_en.md
浏览文件 @ f9413d1a
...@@ -45,6 +45,7 @@ python tools/infer/predict.py \ ...@@ -45,6 +45,7 @@ python tools/infer/predict.py \
- [ResNet50_vc](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar) - [ResNet50_vc](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar)
- [ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar) - [ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar)
- [ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar) - [ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar)
- [ResNet34_vd_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_ssld_pretrained.tar)
- [ResNet50_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar) - [ResNet50_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar)
- [ResNet50_vd_v2](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_v2_pretrained.tar) - [ResNet50_vd_v2](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_v2_pretrained.tar)
- [ResNet101_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_vd_pretrained.tar) - [ResNet101_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_vd_pretrained.tar)
...@@ -149,11 +150,13 @@ python tools/infer/predict.py \ ...@@ -149,11 +150,13 @@ python tools/infer/predict.py \
- HRNet series - HRNet series
- HRNet series<sup>[[13](#ref13)]</sup>([paper link](https://arxiv.org/abs/1908.07919)) - HRNet series<sup>[[13](#ref13)]</sup>([paper link](https://arxiv.org/abs/1908.07919))
- [HRNet_W18_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_pretrained.tar) - [HRNet_W18_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_pretrained.tar)
- [HRNet_W18_C_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_ssld_pretrained.tar)
- [HRNet_W30_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W30_C_pretrained.tar) - [HRNet_W30_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W30_C_pretrained.tar)
- [HRNet_W32_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W32_C_pretrained.tar) - [HRNet_W32_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W32_C_pretrained.tar)
- [HRNet_W40_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar) - [HRNet_W40_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar)
- [HRNet_W44_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar) - [HRNet_W44_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar)
- [HRNet_W48_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar) - [HRNet_W48_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar)
- [HRNet_W48_C_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_ssld_pretrained.tar)
- [HRNet_W64_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar) - [HRNet_W64_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar)
......
docs/en/update_history_en.md
浏览文件 @ f9413d1a
# Release Notes # Release Notes
* 2020.09.17
* Add `HRNet_W48_C_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 83.62%.
* Add `ResNet34_vd_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 79.72%.
* 2020.09.07 * 2020.09.07
* Add `HRNet_W18_C_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 81.16%. * Add `HRNet_W18_C_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 81.16%.
* Add `MobileNetV3_small_x0_35_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 55.55%. * Add `MobileNetV3_small_x0_35_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 55.55%.
...@@ -9,7 +13,7 @@ ...@@ -9,7 +13,7 @@
* Add `Fix_ResNet50_vd_ssld_v2` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 84.00%. * Add `Fix_ResNet50_vd_ssld_v2` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 84.00%.
* 2020.06.17 * 2020.06.17
* Add English documents。 * Add English documents.
* 2020.06.12 * 2020.06.12
* Add support for training and evaluation on Windows or CPU. * Add support for training and evaluation on Windows or CPU.
......
docs/zh_CN/models/HRNet.md
浏览文件 @ f9413d1a
...@@ -27,6 +27,7 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络， ...@@ -27,6 +27,7 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络，
| HRNet_W40_C | 0.788 | 0.945 | 0.789 | 0.945 | 25.410 | 57.550 | | HRNet_W40_C | 0.788 | 0.945 | 0.789 | 0.945 | 25.410 | 57.550 |
| HRNet_W44_C | 0.790 | 0.945 | 0.789 | 0.944 | 29.790 | 67.060 | | HRNet_W44_C | 0.790 | 0.945 | 0.789 | 0.944 | 29.790 | 67.060 |
| HRNet_W48_C | 0.790 | 0.944 | 0.793 | 0.945 | 34.580 | 77.470 | | HRNet_W48_C | 0.790 | 0.944 | 0.793 | 0.945 | 34.580 | 77.470 |
| HRNet_W48_C_ssld | 0.836 | 0.968 | 0.793 | 0.945 | 34.580 | 77.470 |
| HRNet_W64_C | 0.793 | 0.946 | 0.795 | 0.946 | 57.830 | 128.060 | | HRNet_W64_C | 0.793 | 0.946 | 0.795 | 0.946 | 57.830 | 128.060 |
...@@ -41,6 +42,7 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络， ...@@ -41,6 +42,7 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络，
| HRNet_W40_C | 224 | 256 | 10.739 | | HRNet_W40_C | 224 | 256 | 10.739 |
| HRNet_W44_C | 224 | 256 | 11.497 | | HRNet_W44_C | 224 | 256 | 11.497 |
| HRNet_W48_C | 224 | 256 | 12.165 | | HRNet_W48_C | 224 | 256 | 12.165 |
| HRNet_W48_C_ssld | 224 | 256 | 12.165 |
| HRNet_W64_C | 224 | 256 | 15.003 | | HRNet_W64_C | 224 | 256 | 15.003 |
...@@ -57,4 +59,5 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络， ...@@ -57,4 +59,5 @@ HRNet是2019年由微软亚洲研究院提出的一种全新的神经网络，
| HRNet_W40_C | 224 | 256 | 11.4229 | 19.1595 | 30.47984 | 12.12202 | 25.68184 | 48.90623 | | HRNet_W40_C | 224 | 256 | 11.4229 | 19.1595 | 30.47984 | 12.12202 | 25.68184 | 48.90623 |
| HRNet_W44_C | 224 | 256 | 12.25778 | 22.75456 | 32.61275 | 13.19858 | 32.25202 | 59.09871 | | HRNet_W44_C | 224 | 256 | 12.25778 | 22.75456 | 32.61275 | 13.19858 | 32.25202 | 59.09871 |
| HRNet_W48_C | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 | | HRNet_W48_C | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 |
| HRNet_W48_C_ssld | 224 | 256 | 12.65015 | 23.12886 | 33.37859 | 13.70761 | 34.43572 | 63.01219 |
| HRNet_W64_C | 224 | 256 | 15.10428 | 27.68901 | 40.4198 | 17.57527 | 47.9533 | 97.11228 | | HRNet_W64_C | 224 | 256 | 15.10428 | 27.68901 | 40.4198 | 17.57527 | 47.9533 | 97.11228 |
docs/zh_CN/models/ResNet_and_vd.md
浏览文件 @ f9413d1a
...@@ -32,6 +32,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠 ...@@ -32,6 +32,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠
| ResNet18_vd | 0.723 | 0.908 | | | 4.140 | 11.710 | | ResNet18_vd | 0.723 | 0.908 | | | 4.140 | 11.710 |
| ResNet34 | 0.746 | 0.921 | 0.732 | 0.913 | 7.360 | 21.800 | | ResNet34 | 0.746 | 0.921 | 0.732 | 0.913 | 7.360 | 21.800 |
| ResNet34_vd | 0.760 | 0.930 | | | 7.390 | 21.820 | | ResNet34_vd | 0.760 | 0.930 | | | 7.390 | 21.820 |
| ResNet34_vd_ssld | 0.797 | 0.949 | | | 7.390 | 21.820 |
| ResNet50 | 0.765 | 0.930 | 0.760 | 0.930 | 8.190 | 25.560 | | ResNet50 | 0.765 | 0.930 | 0.760 | 0.930 | 8.190 | 25.560 |
| ResNet50_vc | 0.784 | 0.940 | | | 8.670 | 25.580 | | ResNet50_vc | 0.784 | 0.940 | | | 8.670 | 25.580 |
| ResNet50_vd | 0.791 | 0.944 | 0.792 | 0.946 | 8.670 | 25.580 | | ResNet50_vd | 0.791 | 0.944 | 0.792 | 0.946 | 8.670 | 25.580 |
...@@ -58,6 +59,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠 ...@@ -58,6 +59,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠
| ResNet18_vd | 224 | 256 | 1.603 | | ResNet18_vd | 224 | 256 | 1.603 |
| ResNet34 | 224 | 256 | 2.272 | | ResNet34 | 224 | 256 | 2.272 |
| ResNet34_vd | 224 | 256 | 2.343 | | ResNet34_vd | 224 | 256 | 2.343 |
| ResNet34_vd_ssld | 224 | 256 | 2.343 |
| ResNet50 | 224 | 256 | 2.939 | | ResNet50 | 224 | 256 | 2.939 |
| ResNet50_vc | 224 | 256 | 3.041 | | ResNet50_vc | 224 | 256 | 3.041 |
| ResNet50_vd | 224 | 256 | 3.165 | | ResNet50_vd | 224 | 256 | 3.165 |
...@@ -79,6 +81,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠 ...@@ -79,6 +81,7 @@ ResNet系列模型是在2015年提出的，一举在ILSVRC2015比赛中取得冠
| ResNet18_vd | 224 | 256 | 1.39593 | 2.69063 | 3.88267 | 1.54557 | 3.85363 | 6.88121 | | ResNet18_vd | 224 | 256 | 1.39593 | 2.69063 | 3.88267 | 1.54557 | 3.85363 | 6.88121 |
| ResNet34 | 224 | 256 | 2.23092 | 4.10205 | 5.54904 | 2.34957 | 5.89821 | 10.73451 | | ResNet34 | 224 | 256 | 2.23092 | 4.10205 | 5.54904 | 2.34957 | 5.89821 | 10.73451 |
| ResNet34_vd | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 | | ResNet34_vd | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 |
| ResNet34_vd_ssld | 224 | 256 | 2.23992 | 4.22246 | 5.79534 | 2.43427 | 6.22257 | 11.44906 |
| ResNet50 | 224 | 256 | 2.63824 | 4.63802 | 7.02444 | 3.47712 | 7.84421 | 13.90633 | | ResNet50 | 224 | 256 | 2.63824 | 4.63802 | 7.02444 | 3.47712 | 7.84421 | 13.90633 |
| ResNet50_vc | 224 | 256 | 2.67064 | 4.72372 | 7.17204 | 3.52346 | 8.10725 | 14.45577 | | ResNet50_vc | 224 | 256 | 2.67064 | 4.72372 | 7.17204 | 3.52346 | 8.10725 | 14.45577 |
| ResNet50_vd | 224 | 256 | 2.65164 | 4.84109 | 7.46225 | 3.53131 | 8.09057 | 14.45965 | | ResNet50_vd | 224 | 256 | 2.65164 | 4.84109 | 7.46225 | 3.53131 | 8.09057 | 14.45965 |
......
docs/zh_CN/models/models_intro.md
浏览文件 @ f9413d1a
...@@ -45,6 +45,7 @@ python tools/infer/predict.py \ ...@@ -45,6 +45,7 @@ python tools/infer/predict.py \
- [ResNet50_vc](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar) - [ResNet50_vc](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar)
- [ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar) - [ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar)
- [ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar) - [ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar)
- [ResNet34_vd_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_ssld_pretrained.tar)
- [ResNet50_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar) - [ResNet50_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar)
- [ResNet50_vd_v2](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_v2_pretrained.tar) - [ResNet50_vd_v2](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_v2_pretrained.tar)
- [ResNet101_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_vd_pretrained.tar) - [ResNet101_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_vd_pretrained.tar)
...@@ -149,11 +150,13 @@ python tools/infer/predict.py \ ...@@ -149,11 +150,13 @@ python tools/infer/predict.py \
- HRNet系列 - HRNet系列
- HRNet系列<sup>[[13](#ref13)]</sup>([论文地址](https://arxiv.org/abs/1908.07919)) - HRNet系列<sup>[[13](#ref13)]</sup>([论文地址](https://arxiv.org/abs/1908.07919))
- [HRNet_W18_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_pretrained.tar) - [HRNet_W18_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_pretrained.tar)
- [HRNet_W18_C_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_ssld_pretrained.tar)
- [HRNet_W30_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W30_C_pretrained.tar) - [HRNet_W30_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W30_C_pretrained.tar)
- [HRNet_W32_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W32_C_pretrained.tar) - [HRNet_W32_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W32_C_pretrained.tar)
- [HRNet_W40_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar) - [HRNet_W40_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W40_C_pretrained.tar)
- [HRNet_W44_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar) - [HRNet_W44_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W44_C_pretrained.tar)
- [HRNet_W48_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar) - [HRNet_W48_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_pretrained.tar)
- [HRNet_W48_C_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W48_C_ssld_pretrained.tar)
- [HRNet_W64_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar) - [HRNet_W64_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W64_C_pretrained.tar)
......
docs/zh_CN/update_history.md
浏览文件 @ f9413d1a
# 更新日志 # 更新日志
- 2020.09.17
* 添加HRNet_W48_C_ssld模型，在ImageNet上Top-1 Acc可达0.836；添加ResNet34_vd_ssld模型，在ImageNet上Top-1 Acc可达0.797。
* 2020.09.07 * 2020.09.07
* 添加HRNet_W18_C_ssld模型，在ImageNet上Top-1 Acc可达0.81162；添加MobileNetV3_small_x0_35_ssld模型，在ImageNet上Top-1 Acc可达0.5555。 * 添加HRNet_W18_C_ssld模型，在ImageNet上Top-1 Acc可达0.81162；添加MobileNetV3_small_x0_35_ssld模型，在ImageNet上Top-1 Acc可达0.5555。
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或