fix dpn (#430)

* fix dpn

* fix se resnext101 metrics

README.md

@@ -219,7 +219,7 @@ Accuracy and inference time metrics of SEResNeXt and Res2Net series models are s
 | SE_ResNet50_vd            | 0.7952    | 0.9475    | 4.28393               | 10.38846             | 8.67     | 28.09     | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet50_vd_pretrained.tar)            |
 | SE_ResNeXt50_<br>32x4d        | 0.7844    | 0.9396    | 8.74121               | 13.563               | 8.02     | 26.16     | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_32x4d_pretrained.tar)        |
 | SE_ResNeXt50_vd_<br>32x4d     | 0.8024    | 0.9489    | 9.17134               | 14.76192             | 10.76    | 26.28     | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_vd_32x4d_pretrained.tar)     |
-| SE_ResNeXt101_<br>32x4d       | 0.7912    | 0.9420    | 18.82604              | 25.31814             | 15.02    | 46.28     | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar)       |
+| SE_ResNeXt101_<br>32x4d       | 0.7939    | 0.9443    | 18.82604              | 25.31814             | 15.02    | 46.28     | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar)       |
 | SENet154_vd               | 0.8140    | 0.9548    | 53.79794              | 66.31684             | 45.83    | 114.29    | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar)               |
 
 

README_cn.md

@@ -221,7 +221,7 @@ SEResNeXt与Res2Net系列模型的精度、速度指标如下表所示，更多
 | SE_ResNet50_vd            | 0.7952    | 0.9475    | 4.28393               | 10.38846             | 8.67     | 28.09     | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet50_vd_pretrained.tar)            |
 | SE_ResNeXt50_<br>32x4d        | 0.7844    | 0.9396    | 8.74121               | 13.563               | 8.02     | 26.16     | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_32x4d_pretrained.tar)        |
 | SE_ResNeXt50_vd_<br>32x4d     | 0.8024    | 0.9489    | 9.17134               | 14.76192             | 10.76    | 26.28     | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_vd_32x4d_pretrained.tar)     |
-| SE_ResNeXt101_<br>32x4d       | 0.7912    | 0.9420    | 18.82604              | 25.31814             | 15.02    | 46.28     | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar)       |
+| SE_ResNeXt101_<br>32x4d       | 0.7939    | 0.9443    | 18.82604              | 25.31814             | 15.02    | 46.28     | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar)       |
 | SENet154_vd               | 0.8140    | 0.9548    | 53.79794              | 66.31684             | 45.83    | 114.29    | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar)               |
 
 

docs/en/models/SEResNext_and_Res2Net_en.md

@@ -53,7 +53,7 @@ At present, there are a total of 24 pretrained models of the three categories op
 | SE_ResNet50_vd        | 0.795  | 0.948  |                   |                   | 8.670        | 28.090            |
 | SE_ResNeXt50_32x4d    | 0.784  | 0.940  | 0.789             | 0.945             | 8.020        | 26.160            |
 | SE_ResNeXt50_vd_32x4d | 0.802  | 0.949  |                   |                   | 10.760       | 26.280            |
-| SE_ResNeXt101_32x4d   | 0.791  | 0.942  | 0.793             | 0.950             | 15.020       | 46.280            |
+| SE_ResNeXt101_32x4d   | 0.7939  | 0.9443  | 0.793             | 0.950             | 15.020       | 46.280            |
 | SENet154_vd           | 0.814  | 0.955  |                   |                   | 45.830       | 114.290           |
 
 

docs/zh_CN/models/SEResNext_and_Res2Net.md

@@ -52,7 +52,7 @@ Res2Net是2019年提出的一种全新的对ResNet的改进方案，该方案可
 | SE_ResNet50_vd        | 0.795  | 0.948  |                   |                   | 8.670        | 28.090            |
 | SE_ResNeXt50_32x4d    | 0.784  | 0.940  | 0.789             | 0.945             | 8.020        | 26.160            |
 | SE_ResNeXt50_vd_32x4d | 0.802  | 0.949  |                   |                   | 10.760       | 26.280            |
-| SE_ResNeXt101_32x4d   | 0.791  | 0.942  | 0.793             | 0.950             | 15.020       | 46.280            |
+| SE_ResNeXt101_32x4d   | 0.7939  | 0.9443  | 0.793             | 0.950             | 15.020       | 46.280            |
 | SENet154_vd           | 0.814  | 0.955  |                   |                   | 45.830       | 114.290           |
 
 

ppcls/modeling/architectures/dpn.py

@@ -232,7 +232,7 @@ class DPN(nn.Layer):
             num_filters=init_num_filter,
             filter_size=init_filter_size,
             stride=2,
-            pad=1,
+            pad=init_padding,
             act='relu',
             name="conv1")