1.【删除乱码并打标签】删前缀（用户名），删后缀（乱码），加统一标签，并打开excel微调。（输入项为1.单个文件夹地址 2.需要文件夹内加上的标签名）
*注意：此时若要加多个标签，请用英文逗号“,”隔开。
import os
import openpyxl
import re
UNWANTED_UNITS = ["undefined", "皮皮", "zly324"]
# 第一步：删名称
def rename_files(path):
files = [f for f in os.listdir(path) if os.path.isfile(os.path.join(path, f))]
renamed_files = []
counter = 1
for file in files:
filename, ext = os.path.splitext(file)
# 乱码类
if re.search(r'[a-f0-9]{32}', filename) or not '_' in filename:
renamed = f"({counter})"
counter += 1
# AI出图类
else:
parts = re.split(r'[_]+', filename)
parts.pop(0) # 删除第一个单元
# 删除特定的单元
parts = [part for part in parts if part not in UNWANTED_UNITS]
# 删除所有带数字的单元
parts = [part for part in parts if not any(char.isdigit() for char in part)]
# 结尾规则
# 删除UUID风格数字
while parts and re.search(r'^[a-f0-9\-]{32,}$', parts[-1]):
parts.pop(-1)
# 删除长度小于等于4的部分
while parts and len(parts[-1]) <= 4:
parts.pop(-1)
renamed = '_'.join(parts)
renamed_files.append(renamed + ext)
return renamed_files
# 第二步：增名称
def add_prefix(files, prefix):
prefixed_files = [f"{prefix}_{file}" if not file.startswith(prefix) else file for file in files]
# 删除特定的单元
prefixed_files = ['_'.join([part for part in re.split(r'[_]+', name) if part not in UNWANTED_UNITS]) for name in
prefixed_files]
return prefixed_files
# 第三步：创建Excel并自动打开
def create_and_open_excel(files, renamed_files, path):
wb = openpyxl.Workbook()
ws = wb.active
for original, renamed in zip(files, renamed_files):
ws.append([original, renamed])
excel_path = os.path.join(path, os.path.basename(path) + ".xlsx")
wb.save(excel_path)
# 打开Excel文件
os.system(f'start "" "{excel_path}"')
return excel_path
# 第五步：读取Excel并重命名文件
def rename_files_from_excel(path, excel_path):
wb = openpyxl.load_workbook(excel_path)
ws = wb.active
for row in ws.iter_rows(values_only=True):
original_name, new_name = row
target_path = os.path.join(path, new_name)
# 检查原文件是否存在
if os.path.exists(os.path.join(path, original_name)):
# 如果目标文件名已存在，则添加一个编号
counter = 1
base_name, ext = os.path.splitext(new_name)
while os.path.exists(target_path):
new_name = f"{base_name} ({counter}){ext}"
target_path = os.path.join(path, new_name)
counter += 1
os.rename(os.path.join(path, original_name), target_path)
print("重命名完成。")
# 主函数
def main():
path = input("请输入文件夹地址: ")
files = [f for f in os.listdir(path) if os.path.isfile(os.path.join(path, f))]
renamed_files = rename_files(path)
prefix = input("请输入需要批量命名的词: ")
prefixed_files = add_prefix(renamed_files, prefix)
excel_path = create_and_open_excel(files, prefixed_files, path)
print(f"Excel文件已保存为：{excel_path}")
print("请在Excel里微调B列数据，然后保存和关闭Excel文件。完成后按Enter键继续...")
input()
# 重命名文件
rename_files_from_excel(path, excel_path)
if __name__ == "__main__":
main()
2.【处理下划线】统一只保留第一个下划线，删除后面的下划线（输入项为：总地址）
import os
import shutil
def copy_directory(src, dst):
"""复制 src 目录到 dst 目录。"""
try:
shutil.copytree(src, dst)
except FileExistsError:
print(f"备份目录 '{dst}' 已存在。")
def rename_image_files(directory):
"""重命名图片文件，保留第一个下划线，将其他下划线替换为空格，并在需要时添加递增编号。"""
for root, dirs, files in os.walk(directory):
for file in files:
if file.lower().endswith(('.png', '.jpg', '.jpeg', '.gif')):
# 找到第一个下划线的位置
first_underscore = file.find('_')
if first_underscore != -1:
# 保留第一个下划线，将后续下划线替换为空格
before_first_underscore = file[:first_underscore + 1]
after_first_underscore = file[first_underscore + 1:].replace('_', ' ')
new_file_name = before_first_underscore + after_first_underscore
else:
new_file_name = file
if new_file_name != file:
original_file_path = os.path.join(root, file)
new_file_path = os.path.join(root, new_file_name)
increment = 1
# 循环直到找到不冲突的文件名
while os.path.exists(new_file_path):
# 分离文件名和扩展名
file_name, file_extension = os.path.splitext(new_file_name)
# 添加递增编号
new_file_name = f"{file_name} ({increment}){file_extension}"
new_file_path = os.path.join(root, new_file_name)
increment += 1
os.rename(original_file_path, new_file_path)
print(f"文件 {original_file_path} 已重命名为 {new_file_path}")
def main():
# 请求用户输入要处理的目录路径
input_directory = input("请输入要处理的目录路径: ")
# 检查目录是否存在
if not os.path.exists(input_directory):
print(f"指定的目录 {input_directory} 不存在。")
return
# 创建目录的备份
backup_directory = os.path.join(input_directory, "_backup")
print(f"正在创建备份目录: {backup_directory}")
copy_directory(input_directory, backup_directory)
# 在备份目录中重命名图片文件
print("正在重命名备份目录中的图片文件...")
rename_image_files(backup_directory)
print("重命名操作完成。")
if __name__ == "__main__":
main()
3.【逗号转下划线】大地址下图片文件名所有逗号转换为下划线，适用于输入多条标签情况。（输入项为大地址）
import os
def replace_commas_in_filenames_interactive():
"""
交互式地从用户处获取目录路径，然后遍历该路径及其子目录，
并将图片文件名中的逗号替换为下划线。
"""
path = input("请输入目录路径: ")
modified_files = []
# 支持的图片文件扩展名
image_extensions = ['.jpg', '.jpeg', '.png', '.gif', '.bmp', '.tiff', '.webp']
for root, dirs, files in os.walk(path):
for file in files:
if any(file.endswith(ext) for ext in image_extensions):
# 替换逗号为下划线
new_file = file.replace(',', '_').replace('，', '_')
if new_file != file:
os.rename(os.path.join(root, file), os.path.join(root, new_file))
modified_files.append(os.path.join(root, new_file))
return modified_files
# 运行这个函数
modified_files = replace_commas_in_filenames_interactive()
for f in modified_files:
print(f"Modified file: {f}")
4.步骤1-3参照MJ转大模型训练，不同点在于，接下来需要的是：生成左右对称的镜像（防止图片不够）
import os
from PIL import Image
def flip_and_duplicate_image(image_path, output_path):
"""
Flip an image horizontally and save a copy with a suffix.
"""
try:
with Image.open(image_path) as img:
flipped_img = img.transpose(Image.FLIP_LEFT_RIGHT)
flipped_img.save(output_path)
return True
except Exception as e:
print(f"无法处理图像 {image_path}: {e}")
return False
def process_images_in_directory(directory):
"""
Process all images in a given directory and its subdirectories.
"""
total_files = 0
processed_files = 0
# 预先计算总文件数以显示进度
for root, dirs, files in os.walk(directory):
for file in files:
if file.lower().endswith(('.png', '.jpg', '.jpeg', '.gif', '.bmp')):
total_files += 1
for root, dirs, files in os.walk(directory):
for file in files:
if file.lower().endswith(('.png', '.jpg', '.jpeg', '.gif', '.bmp')):
original_path = os.path.join(root, file)
filename, file_extension = os.path.splitext(file)
output_path = os.path.join(root, f"{filename}(1){file_extension}")
if not os.path.exists(output_path): # 避免覆盖已存在的文件
if flip_and_duplicate_image(original_path, output_path):
processed_files += 1
# 打印处理进度
print(f"处理进度: {processed_files}/{total_files} ({(processed_files/total_files)*100:.2f}%)")
def main():
directory = input("请输入要处理的文件夹的路径: ")
if os.path.exists(directory) and os.path.isdir(directory):
process_images_in_directory(directory)
print("图片处理完成。")
else:
print("提供的路径无效或不是一个目录。")
# 运行脚本
main()
5.根据文件名生成文本，并在在文本内去掉多余元素，只保留描述语
import os
import re
def create_txt_from_image():
# 请求用户输入文件夹地址
root_folder = input("请输入图片所在文件夹的完整路径：")
# 判断路径是否存在
if not os.path.exists(root_folder):
print("路径不存在，请检查输入的地址。")
return
# 用于存储创建的txt文件路径的列表
created_txt_files = []
# 使用os.walk遍历文件夹及其所有子文件夹
for folder_path, dirs, files in os.walk(root_folder):
for file in files:
# 检查文件是否为图片（这里我们检查几种常见的图片格式）
if file.endswith(('.jpg', '.png', '.jpeg')):
# 获取不带扩展名的文件名
base_name = os.path.splitext(file)[0]
# 创建同名的txt文件路径
txt_path = os.path.join(folder_path, base_name + '.txt')
created_txt_files.append(txt_path)
# 将图片文件名（不包括后缀）写入到txt文件中
with open(txt_path, 'w', encoding='utf-8') as txt_file:
txt_file.write(base_name)
print("所有图片对应的txt文件已创建完毕。")
# 编译正则表达式以匹配括号及括号内的数字
bracket_pattern = re.compile(r'\(\d+\)')
underscore_pattern = "_" # 不需要正则表达式匹配单个字符
# 遍历所有txt文件，进行内容修改
for txt_file_path in created_txt_files:
with open(txt_file_path, 'r', encoding='utf-8') as file:
content = file.read()
# 删除内容中的括号及括号内的数字
content = bracket_pattern.sub('', content)
# 把下划线替换成英文逗号
content = content.replace(underscore_pattern, ",")
with open(txt_file_path, 'w', encoding='utf-8') as file:
file.write(content)
print("所有txt文件内容中的括号及括号内的数字已删除，并且所有下划线已转换为英文逗号。")
# 运行函数
create_txt_from_image()
标签：files,os,lora,MJ,file,directory,path,txt,数据 From： https://www.cnblogs.com/zly324/p/17845028.html