Update image translation demo (#68)

* update demo

README.md

@@ -10,10 +10,7 @@ PaddleGAN 是一个基于飞桨的生成对抗网络开发工具包.
 
 ### 图片变换
 <div align='center'>
-  <img src='./docs/imgs/A2B.png'>
-</div>
-<div align='center'>
-  <img src='./docs/imgs/B2A.png'>
+  <img src='./docs/imgs/horse2zebra.gif'>
 </div>
 
 ### 妆容迁移

README_en.md

@@ -10,10 +10,7 @@ PaddleGAN is an development kit of Generative Adversarial Network based on Paddl
 
 ### Image Translation
 <div align='center'>
-  <img src='./docs/imgs/A2B.png'>
-</div>
-<div align='center'>
-  <img src='./docs/imgs/B2A.png'>
+  <img src='./docs/imgs/horse2zebra.gif'>
 </div>
 
 

docs/en_US/tutorials/pix2pix_cyclegan.md

@@ -37,9 +37,12 @@
 
 ## 1.3 Results
 
-![](../imgs/horse2zebra.png)
+![](../../imgs/horse2zebra.png)
 
-[model download](TODO)
+## 1.4 模型下载
+| 模型 | 数据集 | 下载地址 |
+|---|---|---|
+| Pix2Pix_cityscapes  | cityscapes | [Pix2Pix_cityscapes](https://paddlegan.bj.bcebos.com/models/Pix2Pix_cityscapes.pdparams)
 
 
 
@@ -49,7 +52,7 @@
 
    CycleGAN uses unpaired pictures for image translation, input two different images with different styles, and automatically perform style transfer. CycleGAN consists of two generators and two discriminators, generator A is inputting images of style A and outputting images of style B, generator B is inputting images of style B and outputting images of style A. The biggest difference between CycleGAN and pix2pix is that CycleGAN can realize image translation without establishing a one-to-one mapping between the source domain and the target domain.
 
-![](../imgs/cyclegan.png)
+![](../../imgs/cyclegan.png)
 
 ## 2.2 How to use
 
@@ -87,9 +90,13 @@
 
 ## 2.3 Results
 
-![](../imgs/A2B.png)
+![](../../imgs/A2B.png)
 
-[model download](TODO)
+## 2.4 模型下载
+| 模型 | 数据集 | 下载地址 |
+|---|---|---|
+| CycleGAN_cityscapes  | cityscapes | [CycleGAN_cityscapes](https://paddlegan.bj.bcebos.com/models/CycleGAN_cityscapes.pdparams) |
+| CycleGAN_horse2zebra | horse2zebra | [CycleGAN_horse2zebra](https://paddlegan.bj.bcebos.com/models/CycleGAN_horse2zebra.pdparams)
 
 
 # References

docs/en_US/tutorials/psgan.md

@@ -10,7 +10,7 @@ This paper is to address the makeup transfer task, which aims to transfer the ma
 
 ## 2. How to use
 ### 2.1 Test
-Pretrained model can be downloaded under following link: [psgan_weight](https://paddlegan.bj.bcebos.com/models/psgan_weight.pkl)
+Pretrained model can be downloaded under following link: [psgan_weight](https://paddlegan.bj.bcebos.com/models/psgan_weight.pdparams)
 
 Running the following command to complete the makeup transfer task. It will geneate the transfered image in the current path when the program running sucessfully.
 
@@ -79,7 +79,7 @@ Notation: In train phase, the `isTrain` value in makeup.yaml file is `True`, but
 
 Model|Dataset|BatchSize|Inference speed|Download
 ---|:--:|:--:|:--:|:--:
-PSGAN|MT-Dataset| 1 | 1.9s/image (GPU:P40) | [model](https://paddlegan.bj.bcebos.com/models/psgan_weight.pkl)
+PSGAN|MT-Dataset| 1 | 1.9s/image (GPU:P40) | [model](https://paddlegan.bj.bcebos.com/models/psgan_weight.pdparams)
 
 ## 3. Result
 ![](../../imgs/makeup_shifter.png)

docs/zh_CN/tutorials/pix2pix_cyclegan.md

@@ -40,7 +40,10 @@
 
 ![](../../imgs/horse2zebra.png)
 
-[模型下载](TODO)
+## 1.4 模型下载
+| 模型 | 数据集 | 下载地址 |
+|---|---|---|
+| Pix2Pix_cityscapes  | cityscapes | [Pix2Pix_cityscapes](https://paddlegan.bj.bcebos.com/models/Pix2Pix_cityscapes.pdparams)
 
 
 # 2 CycleGAN
@@ -88,7 +91,11 @@
 
 ![](../../imgs/A2B.png)
 
-[模型下载](TODO)
+## 2.4 模型下载
+| 模型 | 数据集 | 下载地址 |
+|---|---|---|
+| CycleGAN_cityscapes  | cityscapes | [CycleGAN_cityscapes](https://paddlegan.bj.bcebos.com/models/CycleGAN_cityscapes.pdparams) |
+| CycleGAN_horse2zebra | horse2zebra | [CycleGAN_horse2zebra](https://paddlegan.bj.bcebos.com/models/CycleGAN_horse2zebra.pdparams)
 
 
 # 参考：

docs/zh_CN/tutorials/psgan.md

@@ -10,7 +10,7 @@
 
 ## 2. 使用方法
 ### 2.1 测试
-预训练模型可以从如下地址下载: [psgan_weight](https://paddlegan.bj.bcebos.com/models/psgan_weight.pkl)
+预训练模型可以从如下地址下载: [psgan_weight](https://paddlegan.bj.bcebos.com/models/psgan_weight.pdparams)
 
 运行如下命令，就可以完成妆容迁移，程序运行成功后，会在当前文件夹生成妆容迁移后的图片文件。本项目中提供了原始图片和参考供展示使用，具体命令如下所示：
 
@@ -75,7 +75,7 @@ data
 ### 2.3 模型
 Model|Dataset|BatchSize|Inference speed|Download
 ---|:--:|:--:|:--:|:--:
-PSGAN|MT-Dataset| 1 | 1.9s(GPU:P40) | [model](https://paddlegan.bj.bcebos.com/models/psgan_weight.pkl)
+PSGAN|MT-Dataset| 1 | 1.9s(GPU:P40) | [model](https://paddlegan.bj.bcebos.com/models/psgan_weight.pdparams)
 
 ## 3. 妆容迁移结果展示
 

ppgan/apps/psgan_predictor.py

@@ -53,7 +53,7 @@ def mask2image(mask: np.array, format="HWC"):
     return canvas
 
 
-PS_WEIGHT_URL = "https://paddlegan.bj.bcebos.com/models/psgan_weight.pkl"
+PS_WEIGHT_URL = "https://paddlegan.bj.bcebos.com/models/psgan_weight.pdparams"
 
 
 class PreProcess:

ppgan/engine/trainer.py

@@ -256,7 +256,7 @@ class Trainer:
         assert name in ['checkpoint', 'weight']
 
         state_dicts = {}
-        save_filename = 'epoch_%s_%s.pkl' % (epoch, name)
+        save_filename = 'epoch_%s_%s.pdparams' % (epoch, name)
         save_path = os.path.join(self.output_dir, save_filename)
         for net_name, net in self.model.nets.items():
             state_dicts[net_name] = net.state_dict()
@@ -275,7 +275,8 @@ class Trainer:
         if keep > 0:
             try:
                 checkpoint_name_to_be_removed = os.path.join(
-                    self.output_dir, 'epoch_%s_%s.pkl' % (epoch - keep, name))
+                    self.output_dir,
+                    'epoch_%s_%s.pdparams' % (epoch - keep, name))
                 if os.path.exists(checkpoint_name_to_be_removed):
                     os.remove(checkpoint_name_to_be_removed)