根据文件内容进行文件提取分类

def read_imdb2(data_dir):
    """读取数据集文本序列和标签"""
    data = []
    tag=0
    new_file='F:\\dll1'
    for file in os.listdir(data_dir):
        filenamedir=os.path.join(data_dir, file)
        with open(filenamedir, 'rb') as f:
            # join用来做路径拼接
            review = f.read()
            # print(review[0])
            #根据文件夹的内容选定特征值去判断，然后进行分类，在此利用的是文件开头是"MZ"
            M,Z=review[0],review[1]
            M,Z=chr(M),chr(Z)
            # print(M+Z)
            if M == 'M' and Z == 'Z':
                print(review[0:3])
                new_file1 = os.path.join(new_file, file)
                print(new_file1)
                shutil.copy(filenamedir, new_file1)
                tag=tag+1
    print("共有文件{}".format(tag))