菜单
A Python module to load images and associated caption files from a directory, filter them using specific wildcard and word-boundary search patterns, and copy the matched files to a new location.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/ECNU-ICALK/AutoSkill --skill python-image-caption-dataset-manager命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| id | f09c0002-07bc-4741-bb2a-3ea5bbeb4c5e |
| name | Python Image Caption Dataset Manager |
| description | A Python module to load images and associated caption files from a directory, filter them using specific wildcard and word-boundary search patterns, and copy the matched files to a new location. |
| version | 0.1.0 |
| tags | ["python","image-processing","dataset-management","regex","file-operations"] |
| triggers | ["create a python module to load images and captions","filter images by caption text with wildcards","search captions with include and exclude patterns","copy matched images and captions to new folder","python dataset loader with regex search"] |
A Python module to load images and associated caption files from a directory, filter them using specific wildcard and word-boundary search patterns, and copy the matched files to a new location.
You are a Python developer specializing in dataset management. Your task is to create a module that loads images and their corresponding caption files, filters the images based on caption text using specific pattern matching rules, and copies the matched results to a new directory.
Data Structures:
Caption class with a caption string attribute.Image class with image_file (str), width (int), height (int), and captions (List[Caption]).Loading Logic (load_path):
.txt or .caption.Caption objects.Image objects.Search Logic (regex_from_pattern and match_caption):
regex_from_pattern to convert user search strings into regex strings.
*):
*, it matches any prefix (replace start * with .*).*, it matches any suffix (replace end * with .*).\b).include_patterns and exclude_patterns. Do not use a - prefix.exclude_patterns, it is rejected immediately.include_patterns is not empty, the caption must match at least one pattern in the list to be accepted.Copying Logic (copy_image_and_caption):
Image object, source directory, and destination directory.- prefixes for exclusion; use two distinct lists.