1、当数据集是压缩包时，需要解压成图片文件

import zipfile
def unzip_data(src_path,target_path):
    '''
    解压原始数据集，将src_path路径下的zip包解压至data/dataset目录下
    '''
    if(not os.path.isdir(target_path)):    
        z = zipfile.ZipFile(src_path, 'r')
        z.extractall(path=target_path)
        z.close()
    else:
        print("文件已解压")

2、生成json文件

# 说明的json文件信息
    readjson = {}
    readjson['all_class_name'] = data_list_path                  #文件父目录
    readjson['all_class_images'] = all_class_images
    readjson['class_detail'] = class_detail
    jsons = json.dumps(readjson, sort_keys=True, indent=4, separators=(',', ': '))
    with open(train_parameters['readme_path'],'w') as f:
        f.write(jsons)
    print ('生成数据列表完成！')

3、自定义data_reader

def data_reader(file_list):
    '''
    自定义data_reader
    '''
    def reader():
        with open(file_list, 'r') as f:
            lines = [line.strip() for line in f]
            for line in lines:
                img_path, lab = line.strip().split('\t')
                img = cv2.imread(img_path)
                img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
                img = np.array(img).astype('float32')
                img = img/255.0
                yield img, int(lab) 
    return reader