add cuda/cudnn version.

fluid/PaddleCV/video/README.md

@@ -44,7 +44,7 @@ Paddle提供默认配置文件位于`./configs`文件夹下，五种模型对应
 4. stnet.txt
 5. tsn.txt
 
-详细使用步骤请参考各模型文档
+详细使用步骤请参考各模型文档：[Attention Cluster](./models/attention_cluster/README.md), [Attention LSTM](./models/attention_lstm/README.md), [NeXtVLAD](./models/nextvlad/README.md), [StNet](./models/stnet/README.md), [TSN](./models/tsn/README.md)
 
 ## 模型精度
 
@@ -60,11 +60,11 @@ Paddle提供默认配置文件位于`./configs`文件夹下，五种模型对应
 
 ## Model Zoo
 
-| 模型 | Batch Size | 环境配置 | 下载链接 |
-| :-------: | :---: | :---------: | :----------: |
-| Attention Cluster | 2048 | 8卡P40 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_cluster_youtube8m.tar.gz) |
-| Attention LSTM | 1024 | 8卡P40 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_lstm_youtube8m.tar.gz) |
-| NeXtVLAD | 160 | 4卡P40 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_cluster_youtube8m.tar.gz) |
-| StNet | 128 | 8卡P40 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/stnet_kientics.tar.gz) |
-| TSN | 256 | 8卡P40 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/tsn_kientics.tar.gz) |
+| 模型 | Batch Size | 环境配置 | CUDA版本 | CUDNN版本 | 下载链接 |
+| :-------: | :---: | :---------: | :----: | :-----: | :----------: |
+| Attention Cluster | 2048 | 8卡P40 | 8.0 | 7.1 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_cluster_youtube8m.tar.gz) |
+| Attention LSTM | 1024 | 8卡P40 | 8.0 | 7.1 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_lstm_youtube8m.tar.gz) |
+| NeXtVLAD | 160 | 4卡P40 | 8.0 | 7.1 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/attention_cluster_youtube8m.tar.gz) |
+| StNet | 128 | 8卡P40 | 8.0 | 5.1 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/stnet_kientics.tar.gz) |
+| TSN | 256 | 8卡P40 | 8.0 | 7.1 | [model](https://paddlemodels.bj.bcebos.com/video_clasification/tsn_kientics.tar.gz) |
 

fluid/PaddleCV/video/models/attention_cluster/README.md

@@ -100,7 +100,7 @@ Attention Cluster模型使用2nd-Youtube-8M数据集, 数据下载及准备请
 
 ## 参考论文
 
-- [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550) Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
+- [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550), Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
 
 ## 版本更新
 

fluid/PaddleCV/video/models/attention_lstm/README.md

@@ -93,7 +93,7 @@ AttentionLSTM模型使用2nd-Youtube-8M数据集，关于数据本分请参考[
 
 - [Beyond Short Snippets: Deep Networks for Video Classification](https://arxiv.org/abs/1503.08909) Joe Yue-Hei Ng, Matthew Hausknecht, Sudheendra Vijayanarasimhan, Oriol Vinyals, Rajat Monga, George Toderici
 
-- [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550) Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
+- [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550), Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
 
 ## 版本更新
 

fluid/PaddleCV/video/models/tsn/README.md

@@ -80,7 +80,7 @@ TSN的训练数据采用由DeepMind公布的Kinetics-400动作识别数据集。
 
 ## 参考论文
 
-- [StNet:Local and Global Spatial-Temporal Modeling for Human Action Recognition](https://arxiv.org/abs/1608.00859) Limin Wang, Yuanjun Xiong, Zhe Wang, Yu Qiao, Dahua Lin, Xiaoou Tang, Luc Van Gool
+- [StNet:Local and Global Spatial-Temporal Modeling for Human Action Recognition](https://arxiv.org/abs/1608.00859), Limin Wang, Yuanjun Xiong, Zhe Wang, Yu Qiao, Dahua Lin, Xiaoou Tang, Luc Van Gool
 
 ## 版本更新