PaddlePaddle · wangna11BD · May 9, 2022 · Apr 15, 2022 · Apr 15, 2022 · Apr 15, 2022
diff --git a/README.md b/README.md
@@ -34,7 +34,7 @@ GAN-Generative Adversarial Network, was praised by "the Father of Convolutional
   <div align='center'>
     <img src='https://user-images.githubusercontent.com/48054808/144848981-00c6ad21-0702-4381-9544-becb227ed9f0.gif' width='300'/>
   </div>
-  
+
 - 😍 **Boy or Girl？：[StyleGAN V2 Face Editing](./docs/en_US/tutorials/styleganv2editing.md)-Changing genders！** 😍
   - **[Online Toturials](https://aistudio.baidu.com/aistudio/projectdetail/2565277?contributionType=1)**
   <div align='center'>
@@ -118,6 +118,7 @@ GAN-Generative Adversarial Network, was praised by "the Father of Convolutional
 * [StarGANv2](docs/en_US/tutorials/starganv2.md)
 * [MPR Net](./docs/en_US/tutorials/mpr_net.md)
 * [FaceEnhancement](./docs/en_US/tutorials/face_enhancement.md)
+* [PReNet](./docs/en_US/tutorials/prenet.md)
 
 
 ## Composite Application

diff --git a/README_cn.md b/README_cn.md
@@ -18,7 +18,7 @@ GAN--生成对抗网络，被“卷积网络之父”**Yann LeCun（杨立昆）
 
 ## 近期活动🔥🔥🔥
 
-  
+
 - 🔥**2021.12.08**🔥
   **💙 AI快车道👩‍🏫：视频超分算法及行业应用 💙**
   - **课程回放链接🔗：https://aistudio.baidu.com/aistudio/education/group/info/25179**
@@ -140,6 +140,7 @@ GAN--生成对抗网络，被“卷积网络之父”**Yann LeCun（杨立昆）
 * 图像视频修复
   * 图像去模糊去噪去雨：[MPR Net](./docs/zh_CN/tutorials/mpr_net.md)
   * 视频去模糊：[EDVR](./docs/zh_CN/tutorials/video_super_resolution.md)
+  * 图像去雨：[PReNet](./docs/zh_CN/tutorials/prenet.md)
 
 ## 产业级应用
 

diff --git a/configs/prenet.yaml b/configs/prenet.yaml
@@ -0,0 +1,98 @@
+total_iters: 300000
+output_dir: output_dir
+find_unused_parameters: True
+checkpoints_dir: checkpoints
+use_dataset: True
+# tensor range for function tensor2img
+min_max:
+  (0., 1.)
+
+model:
+  name: PReNetModel
+  generator:
+    name: PReNet
+  pixel_criterion:
+    name: SSIM
+
+dataset:
+  train:
+    name: SRDataset
+    gt_folder: data/RainTrainH/norain
+    lq_folder: data/RainTrainH/rain
+    num_workers: 4
+    batch_size: 16
+    scale: 1
+    preprocess:
+      - name: LoadImageFromFile
+        key: lq
+      - name: LoadImageFromFile
+        key: gt
+      - name: Transforms
+        input_keys: [lq, gt]
+        pipeline:
+          - name: PairedRandomHorizontalFlip
+            keys: [image, image]
+          - name: PairedRandomVerticalFlip
+            keys: [image, image]
+          - name: PairedRandomTransposeHW
+            keys: [image, image]
+          - name: PairedRandomCrop
+            size: [100, 100]
+            keys: [image, image]
+          - name: PairedToTensor
+            keys: [image, image]
+  test:
+    name: SRDataset
+    gt_folder: data/Rain100H/norain
+    lq_folder: data/Rain100H/rain
+    scale: 1
+    preprocess:
+      - name: LoadImageFromFile
+        key: lq
+      - name: LoadImageFromFile
+        key: gt
+      - name: Transforms
+        input_keys: [lq, gt]
+        pipeline:
+          - name: PairedToTensor
+            keys: [image, image]
+
+
+lr_scheduler:
+  name: MultiStepDecay
+  learning_rate: 0.0013
+  milestones: [36000,60000,96000]
+  gamma: 0.2
+
+optimizer:
+  name: Adam
+  # add parameters of net_name to optim
+  # name should in self.nets
+  net_names:
+    - generator
+  beta1: 0.9
+  beta2: 0.99
+
+validate:
+  interval: 5000
+  save_img: false
+
+  metrics:
+    psnr: # metric name, can be arbitrary
+      name: PSNR
+      crop_border: 0
+      test_y_channel: True
+    ssim:
+      name: SSIM
+      crop_border: 0
+      test_y_channel: True
+
+log_config:
+  interval: 100
+  visiual_interval: 500
+
+snapshot_config:
+  interval: 5000
+
+export_model:
+  - {name: 'generator', inputs_num: 1}
diff --git a/docs/en_US/tutorials/prenet.md b/docs/en_US/tutorials/prenet.md
@@ -0,0 +1,92 @@
+# PReNet
+
+## 1 Introduction
+"Progressive Image Deraining Networks: A Better and Simpler Baseline" provides a better and simpler baseline deraining network by considering network architecture, input and output, and loss functions.
+
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/net.jpg" width=800">
+</div>
+
+## 2 How to use
+
+### 2.1 Prepare dataset
+
+  The dataset(RainH.zip) used by PReNet can be downloaded from [here](https://pan.baidu.com/s/1_vxCatOV3sOA6Vkx1l23eA?pwd=vitu),uncompress it and get two folders(RainTrainH、Rain100H).
+
+  The structure of dataset is as following:
+
+```
+    ├── data
+        ├── RainTrainH
+            ├── rain
+                ├── 1.png
+                └── 2.png
+                    .
+                    .
+            └── norain
+                ├── 1.png
+                └── 2.png
+                    .
+                    .
+        └── Rain100H
+            ├── rain
+                ├── 001.png
+                └── 002.png
+                    .
+                    .
+            └── norain
+                ├── 001.png
+                └── 002.png
+                    .
+                    .
+```
+
+
+### 2.2 Train/Test
+
+
+  train model:
+  ```
+     python -u tools/main.py --config-file configs/prenet.yaml
+  ```
+
+  test model:
+  ```
+     python tools/main.py --config-file configs/prenet.yaml --evaluate-only --load ${PATH_OF_WEIGHT}
+  ```
+
+## 3 Results
+
+Input:
+
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/rain-001.png" width=300">
+</div>
+
+Output:
+
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/derain-rain-001.png" width=300">
+</div>
+
+## 4 Model Download
+| model | dataset |
+|---|---|
+| [PReNet](https://paddlegan.bj.bcebos.com/models/PReNet.pdparams)  | [RainH.zip](https://pan.baidu.com/s/1_vxCatOV3sOA6Vkx1l23eA?pwd=vitu) |
+
+
+
+
+# References
+
+- 1. [Progressive Image Deraining Networks: A Better and Simpler Baseline](https://arxiv.org/pdf/1901.09221v3.pdf)
+
+
+```
+@inproceedings{ren2019progressive,
+   title={Progressive Image Deraining Networks: A Better and Simpler Baseline},
+   author={Ren, Dongwei and Zuo, Wangmeng and Hu, Qinghua and Zhu, Pengfei and Meng, Deyu},
+   booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
+   year={2019},
+ }
+```
diff --git a/docs/zh_CN/tutorials/prenet.md b/docs/zh_CN/tutorials/prenet.md
@@ -0,0 +1,90 @@
+# PReNet
+
+## 1 简介
+Progressive Image Deraining Networks: A Better and Simpler Baseline提出一种多阶段渐进的残差网络，每一个阶段都是resnet，每一res块的输入为上一res块输出和原始雨图，另外采用使用SSIM损失进行训练,进一步提升了网络的性能，网络总体简洁高效，在各种数据集上表现良好,为图像去雨提供了一个很好的基准。
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/net.jpg" width=800">
+</div>
+
+## 2 如何使用
+
+### 2.1 数据准备
+
+  数据集(RainH.zip) 可以在[此处](https://pan.baidu.com/s/1_vxCatOV3sOA6Vkx1l23eA?pwd=vitu)下载,将其解压到./data路径下。
+
+  数据集文件结构如下:
+
+  ```
+    ├── data
+        ├── RainTrainH
+            ├── rain
+                ├── 1.png
+                └── 2.png
+                    .
+                    .
+            └── norain
+                ├── 1.png
+                └── 2.png
+                    .
+                    .
+        └── Rain100H
+            ├── rain
+                ├── 001.png
+                └── 002.png
+                    .
+                    .
+            └── norain
+                ├── 001.png
+                └── 002.png
+                    .
+                    .
+  ```
+
+### 2.2 训练和测试
+
+
+  训练模型：
+  ```
+     python -u tools/main.py --config-file configs/prenet.yaml
+  ```
+
+  测试模型：
+  ```
+     python tools/main.py --config-file configs/prenet.yaml --evaluate-only --load ${PATH_OF_WEIGHT}
+  ```
+
+## 3 预测结果
+
+输入：
+
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/rain-001.png" width=300">
+</div>
+
+输出：
+
+<div align="center">
+    <img src="https://github.com/simonsLiang/PReNet_paddle/blob/main/data/derain-rain-001.png" width=300">
+</div>
+
+## 4 模型参数下载
+| 模型 | 数据集 |
+|---|---|
+| [PReNet](https://paddlegan.bj.bcebos.com/models/PReNet.pdparams)  | [RainH.zip](https://pan.baidu.com/s/1_vxCatOV3sOA6Vkx1l23eA?pwd=vitu) |
+
+
+
+
+## 参考
+
+- 1. [Progressive Image Deraining Networks: A Better and Simpler Baseline](https://arxiv.org/pdf/1901.09221v3.pdf)
+
+
+```
+@inproceedings{ren2019progressive,
+   title={Progressive Image Deraining Networks: A Better and Simpler Baseline},
+   author={Ren, Dongwei and Zuo, Wangmeng and Hu, Qinghua and Zhu, Pengfei and Meng, Deyu},
+   booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
+   year={2019},
+ }
+```
diff --git a/ppgan/datasets/preprocess/__init__.py b/ppgan/datasets/preprocess/__init__.py
@@ -3,6 +3,6 @@
                          PairedRandomVerticalFlip, PairedRandomTransposeHW,
                          SRPairedRandomCrop, SplitPairedImage, SRNoise,
                          NormalizeSequence, MirrorVideoSequence,
-                         TransposeSequence)
+                         TransposeSequence, PairedToTensor)
 
 from .builder import build_preprocess