Linux下Pytorch版deeplabv3+环境配置训练自己的数据集

ゞ浴缸里的玫瑰 2022-12-23 07:19 258阅读 0赞

### Linux下Pytorch版deeplabv3+环境配置训练自己的数据集 ###

*   *  开发环境
     *  数据集准备
     *   *  1.VOC数据集格式
         *   *  \`JPEGImages\`里面放原图
             *  \`SegmentationClass\`里面放对应的mask图片png格式，注意要和\`JPEGImages\`里的图片一一对应
             *  \`ImageSets/ Segmentation\`的txt文件中放去掉后缀的图片名
         *  2.json转换mask图片
         *  3.提取出所有文件夹中的\`label.png\`并改成对应的名字放在指定目录中
     *  修改代码
     *   *  1.在\`mypath.py\`中添加自己的数据集名称与路径
         *  2.在同级目录中修改\`train.py\`约185行添加自己数据集的名称(可以设置为默认)
         *  3.在dataloaders目录下修改\_\_init\_\_.py
         *  4. 修改dateloaders目录下\`utils.py\`
         *  5.在dataloaders/datasets目录下添加文件
     *  运行并测试
     *   *  1.开始训练
         *  2.测试

## 开发环境 ##

我使用的是实验室服务器的环境：

*  **CentOS**(windows也可)
 *  **pytorch-gpu1.7**
 *  **python3.7**

代码在这：[链接][Link 1]

## 数据集准备 ##

### 1.VOC数据集格式 ###

文件安排如下：

- ImageSets
    	- Segmentation
    		- train.txt
    		- trainval.txt
    		- val.txt
    - JPEGImages
    - SegmentationClass

#### `JPEGImages`里面放原图 ####

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center]

#### `SegmentationClass`里面放对应的mask图片png格式，注意要和`JPEGImages`里的图片一一对应 ####

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 1]

#### `ImageSets/ Segmentation`的txt文件中放去掉后缀的图片名 ####

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 2]  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 3]

### 2.json转换mask图片 ###

数据转换参考的[这里][Link 2]

*  用下面的代码替换labelme环境下的对应文件`Anaconda3\envs\labelme\Lib\site-packages\labelme\cli`（注意替换掉`path`的路径，换上自己的json文件路径）

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 4]

import argparse
    import base64
    import json
    import os
    import os.path as osp
    
    import PIL.Image
    import yaml
    
    from labelme.logger import logger
    from labelme import utils
    
    path = "G:/Seg552VOC/seg552Json"
    dirs = os.listdir(path)
    
    
    def label(json_file, out_dir, label_name_to_value):
        data = json.load(open(json_file))
        if data['imageData']:
            imageData = data['imageData']
        else:
            imagePath = os.path.join(os.path.dirname(json_file), data['imagePath'])
            with open(imagePath, 'rb') as f:
                imageData = f.read()
                imageData = base64.b64encode(imageData).decode('utf-8')
        img = utils.img_b64_to_arr(imageData)
    
        for shape in sorted(data['shapes'], key=lambda x: x['label']):
            label_name = shape['label']
            if label_name in label_name_to_value:
                label_value = label_name_to_value[label_name]
            else:
                label_value = len(label_name_to_value)
                label_name_to_value[label_name] = label_value
        lbl = utils.shapes_to_label(img.shape, data['shapes'], label_name_to_value)
    
        label_names = [None] * (max(label_name_to_value.values()) + 1)
        for name, value in label_name_to_value.items():
            label_names[value] = name
        lbl_viz = utils.draw_label(lbl, img, label_names)
    
        PIL.Image.fromarray(img).save(osp.join(out_dir, 'img.png'))
        utils.lblsave(osp.join(out_dir, 'label.png'), lbl)
        PIL.Image.fromarray(lbl_viz).save(osp.join(out_dir, 'label_viz.png'))
    
        with open(osp.join(out_dir, 'label_names.txt'), 'w') as f:
            for lbl_name in label_names:
                f.write(lbl_name + '\n')
    
        logger.warning('info.yaml is being replaced by label_names.txt')
        info = dict(label_names=label_names)
        with open(osp.join(out_dir, 'info.yaml'), 'w') as f:
            yaml.safe_dump(info, f, default_flow_style=False)
    
        logger.info('Saved to: {}'.format(out_dir))
    
    
    def main():
        logger.warning('This script is aimed to demonstrate how to convert the'
                       'JSON file to a single image dataset, and not to handle'
                       'multiple JSON files to generate a real-use dataset.')
    
        parser = argparse.ArgumentParser()
        parser.add_argument('json_file_dir')
        parser.add_argument('-o', '--out', default=None)
        args = parser.parse_args()
        label_name_to_value = { '_background_': 0}
    
        for json_file in dirs:
            if args.out is None:
                out_dir = osp.basename(json_file).replace('.', '_')
                out_dir = osp.join(osp.dirname(json_file), out_dir)
            else:
                out_dir = args.out
            if not osp.exists(out_dir):
                os.mkdir(out_dir)
    
            label(json_file, out_dir, label_name_to_value)
    if __name__ == '__main__':
        main()

*  在conda prompt中启动labelme环境运行`labelme_json_to_dataset 输出的目录`，得到很多文件夹，每个文件夹里都有四个文件，我们只需要`label.png`  
    ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 5]  
    ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 6]

### 3.提取出所有文件夹中的`label.png`并改成对应的名字放在指定目录中 ###

`label.py`代码如下：

import os
    import shutil
    
    inputdir = 'G:/Seg552VOC/datasetDir'
    outputdir = 'G:/Seg552VOC/ImageSetNew'
    
    for dir in os.listdir(inputdir):
        # 设置旧文件名（就是路径+文件名）
        oldname = inputdir + os.sep + dir + os.sep + 'label.png'  # os.sep添加系统分隔符
    
        # 设置新文件名
        newname = outputdir + os.sep + dir.split('_')[0] + '.png'
    
        shutil.copyfile(oldname, newname)  # 用os模块中的rename方法对文件改名
        print(oldname, '======>', newname)

*  得到下面的文件，就是我们所需要的`SegmentationClass`  
    ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 7]

## 修改代码 ##

参考博客：[链接][Link 3]

### 1.在`mypath.py`中添加自己的数据集名称与路径 ###

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 8]

### 2.在同级目录中修改`train.py`约185行添加自己数据集的名称(可以设置为默认) ###

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 9]

### 3.在dataloaders目录下修改\_\_init\_\_.py ###

在第一行添加数据集名称，复制`'pascal'`数据集描述，把名称修改为自己数据集的名字  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 10]

### 4. 修改dateloaders目录下`utils.py` ###

在76行左右添加代码，设置每一类别的颜色显示。  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 11]

在24行左右添加代码，其中`n_classes`是你要分割的类别数  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 12]

### 5.在dataloaders/datasets目录下添加文件 ###

复制一份`pascal.py`，并重命名文件  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 13]  
修改里面的类别数和数据集名称  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 14]

## 运行并测试 ##

### 1.开始训练 ###

运行指令如下

python train.py --backbone mobilenet --lr 0.007 --workers 1 --epochs 50 --batch-size 8 --gpu-ids 0 --checkname deeplab-mobilenet

–backbone mobilenet 指的是使用mobilenet作为backbone  
–gpu-ids 0 指定gpu  
–checkname deeplab-mobilenet 使用mobilenet预训练模型  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 15]

### 2.测试 ###

测试testdemo.py  
修改–in-path为数据集的测试图片，最后的结果保存在–in-path中

#
    # demo.py
    #
    import argparse
    import os
    import numpy as np
    import time
    
    from modeling.deeplab import *
    from dataloaders import custom_transforms as tr
    from PIL import Image
    from torchvision import transforms
    from dataloaders.utils import  *
    from torchvision.utils import make_grid, save_image
    
    def main():
    
        parser = argparse.ArgumentParser(description="PyTorch DeeplabV3Plus Training")
        parser.add_argument('--in-path', type=str,  default='/root/home/zyx/Seg552_VOC/test',
                            help='image to test')
        # parser.add_argument('--out-path', type=str, required=True, help='mask image to save')
        parser.add_argument('--backbone', type=str, default='resnet',
                            choices=['resnet', 'xception', 'drn', 'mobilenet'],
                            help='backbone name (default: resnet)')
        parser.add_argument('--ckpt', type=str, default='deeplab-resnet.pth',
                            help='saved model')
        parser.add_argument('--out-stride', type=int, default=16,
                            help='network output stride (default: 8)')
        parser.add_argument('--no-cuda', action='store_true', default=False,
                            help='disables CUDA training')
        parser.add_argument('--gpu-ids', type=str, default='0',
                            help='use which gpu to train, must be a \
                            comma-separated list of integers only (default=0)')
        parser.add_argument('--dataset', type=str, default='pascal',
                            choices=['pascal', 'coco', 'cityscapes','invoice'],
                            help='dataset name (default: pascal)')
        parser.add_argument('--crop-size', type=int, default=513,
                            help='crop image size')
        parser.add_argument('--num_classes', type=int, default=4,
                            help='crop image size')
        parser.add_argument('--sync-bn', type=bool, default=None,
                            help='whether to use sync bn (default: auto)')
        parser.add_argument('--freeze-bn', type=bool, default=False,
                            help='whether to freeze bn parameters (default: False)')
    
        args = parser.parse_args()
        args.cuda = not args.no_cuda and torch.cuda.is_available()
        if args.cuda:
            try:
                args.gpu_ids = [int(s) for s in args.gpu_ids.split(',')]
            except ValueError:
                raise ValueError('Argument --gpu_ids must be a comma-separated list of integers only')
    
        if args.sync_bn is None:
            if args.cuda and len(args.gpu_ids) > 1:
                args.sync_bn = True
            else:
                args.sync_bn = False
        model_s_time = time.time()
        model = DeepLab(num_classes=args.num_classes,
                        backbone=args.backbone,
                        output_stride=args.out_stride,
                        sync_bn=args.sync_bn,
                        freeze_bn=args.freeze_bn)
    
        ckpt = torch.load(args.ckpt, map_location='cpu')
        model.load_state_dict(ckpt['state_dict'])
        model = model.cuda()
        model_u_time = time.time()
        model_load_time = model_u_time-model_s_time
        print("model load time is {}".format(model_load_time))
    
        composed_transforms = transforms.Compose([
            tr.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
            tr.ToTensor()])
        for name in os.listdir(args.in_path):
            s_time = time.time()
            image = Image.open(args.in_path+"/"+name).convert('RGB')
    
            # image = Image.open(args.in_path).convert('RGB')
            target = Image.open(args.in_path+"/"+name).convert('L')
            sample = { 'image': image, 'label': target}
            tensor_in = composed_transforms(sample)['image'].unsqueeze(0)
    
            model.eval()
            if args.cuda:
                tensor_in = tensor_in.cuda()
            with torch.no_grad():
                output = model(tensor_in)
    
            grid_image = make_grid(decode_seg_map_sequence(torch.max(output[:3], 1)[1].detach().cpu().numpy()),
                                    3, normalize=False, range=(0, 255))
            save_image(grid_image,args.in_path+"/"+"{}_mask.png".format(name[0:-4]))
            u_time = time.time()
            img_time = u_time-s_time
            print("image:{} time: {} ".format(name,img_time))
            # save_image(grid_image, args.out_path)
            # print("type(grid) is: ", type(grid_image))
            # print("grid_image.shape is: ", grid_image.shape)
        print("image save in in_path.")
    if __name__ == "__main__":
       main()
    
    # python demo.py --in-path your_file --out-path your_dst_file

测试命令

python testdemo.py --ckpt run/Seg552/deeplab-mobilenet/checkpoint.pth.tar --backbone mobilenet

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 16]

参考博客:  
[Pytorch 语义分割DeepLabV3+ 训练自己的数据集 从数据准备到模型训练][Pytorch _DeepLabV3_ _]  
[【windows10】使用pytorch版本deeplabv3+训练自己数据集][windows10_pytorch_deeplabv3]  
[制作自己的语义分割数据集（仿voc2012数据集）][Link 2]

[Link 1]: https://github.com/jfzhang95/pytorch-deeplab-xception
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center]: /images/20221120/91c45c82b553460aabf90c28f1acb1df.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 1]: /images/20221120/b753ef39b601409089964f5d30f60ed0.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 2]: /images/20221120/48aace2d4d8141048e729369aa657db1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 3]: /images/20221120/9aac8ff4e2f742b39e6ed809f399bcc5.png
[Link 2]: https://blog.csdn.net/qq_43631789/article/details/102700231?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-3.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-3.control
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 4]: /images/20221120/77ba569ae1014b34ac63d9b4a453b535.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 5]: /images/20221120/77cdb8918fda4409ba16b6529469fcad.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 6]: /images/20221120/9d53a9560e0e4cd5bbf6d20a9a40beea.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 7]: /images/20221120/84a0bdd0aa81464e95947813159e016f.png
[Link 3]: https://blog.csdn.net/qq_39056987/article/details/106455828?utm_medium=distribute.pc_relevant_download.none-task-blog-baidujs-1.nonecase&depth_1-utm_source=distribute.pc_relevant_download.none-task-blog-baidujs-1.nonecase
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 8]: /images/20221120/63a2251639dc43b9ac88515951220895.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 9]: /images/20221120/ae5a47b048b24de5a09c1cb7f2039082.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 10]: /images/20221120/889d61e80b4f449985ec5b5ce036ef70.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 11]: /images/20221120/ce529b9f0d9d4f41b09f967dd937436f.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 12]: /images/20221120/6a9b069e95184237833693b808e9cb9c.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 13]: /images/20221120/ba4957eccbb943628b3f48d0008169db.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 14]: /images/20221120/b5f30e29541541249d9c09bdd621cbfc.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 15]: /images/20221120/443779539686461fa52ebc9719a0bb86.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM2NzY2NTYw_size_16_color_FFFFFF_t_70_pic_center 16]: /images/20221120/ad41ba593eb24dc4bb5e5ce094917aee.png
[Pytorch _DeepLabV3_ _]: https://blog.csdn.net/weixin_41919571/article/details/107906066
[windows10_pytorch_deeplabv3]: https://blog.csdn.net/qq_39056987/article/details/106455828?utm_medium=distribute.pc_relevant.none-task-blog-title-2&spm=1001.2101.3001.4242