用Python代码实现数据集的批量处理（一）

一：解决数据集命名重复的问题

把第一位的0改成c

jpg版本

import os

def rename_files_in_directory(directory):
    # 遍历目录下所有文件
    for filename in os.listdir(directory):
        if filename.endswith('.jpg'):
            filepath = os.path.join(directory, filename)
            # 获取文件名
            file_name, file_extension = os.path.splitext(filename)
            # 如果文件名以数字0开头，则进行替换
            if file_name[0] == '0':
                new_file_name = 'c' + file_name[1:] + file_extension
                # 重命名文件
                os.rename(filepath, os.path.join(directory, new_file_name))

# 指定目录
directory = 'D:/Desktop/datasets_car/images/val'

# 调用函数
rename_files_in_directory(directory)

txt版本 

import os

def rename_files_in_directory(directory):
    # 遍历目录下所有文件
    for filename in os.listdir(directory):
        if filename.endswith('.txt'):
            filepath = os.path.join(directory, filename)
            # 获取文件名
            file_name, file_extension = os.path.splitext(filename)
            # 如果文件名以数字0开头，则进行替换
            if file_name[0] == '0':
                new_file_name = 'c' + file_name[1:] + file_extension
                # 重命名文件
                os.rename(filepath, os.path.join(directory, new_file_name))

# 指定目录
directory = 'D:/Desktop/datasets_car/labels/val'

# 调用函数
rename_files_in_directory(directory)

二： 在原有的数据集上添加新的类别

原来是只有一类0，加在新的数据集上的12类之后，改成12类。

import os

def replace_first_zero_with_12_in_txt_files(directory):
    # 遍历目录下所有文件
    for filename in os.listdir(directory):
        if filename.endswith('.txt'):
            filepath = os.path.join(directory, filename)
            # 读取文件内容
            with open(filepath, 'r') as file:
                lines = file.readlines()
            # 修改文件内容
            with open(filepath, 'w') as file:
                for line in lines:
                    # 利用split方法分割每行字符串，取第一个元素进行替换
                    parts = line.strip().split(' ')
                    if parts[0] == '0':
                        parts[0] = '12'
                    # 重新将修改后的行写入文件
                    file.write(' '.join(parts) + '\n')

# 指定目录
directory = 'D:/Desktop/datasets_car/labels/val'

# 调用函数
replace_first_zero_with_12_in_txt_files(directory)

三：文件后缀.txt转换成.c文件 

import os
import chardet

def detect_encoding(file_path):
    with open(file_path, 'rb') as f:
        result = chardet.detect(f.read())
    return result['encoding']

def txt_to_c(txt_file, c_file):
    encoding = detect_encoding(txt_file)
    with open(txt_file, 'r', encoding=encoding) as f:
        txt_content = f.read()

    with open(c_file, 'w', encoding='utf-8') as f:
        f.write(txt_content)

def batch_convert(txt_folder, c_folder):
    if not os.path.exists(c_folder):
        os.makedirs(c_folder)

    for filename in os.listdir(txt_folder):
        if filename.endswith('.txt'):
            txt_file = os.path.join(txt_folder, filename)
            c_file = os.path.join(c_folder, filename.replace('.txt', '.c'))
            txt_to_c(txt_file, c_file)

if __name__ == "__main__":
    txt_folder = "D:/Desktop/input_directory"  # 原始txt文件所在的文件夹
    c_folder = "D:/Desktop/output_directory"  # 保存转换后c文件的文件夹

    batch_convert(txt_folder, c_folder)