Can I use it with existing notebook code?

Yes, the implementation includes a backward compatibility layer that maps the new structured paths to legacy variable names, ensuring existing processing scripts still function.

How are notebook templates handled?

The skill copies template notebooks from the core repository to the user's project folder, allowing users to modify their local copies freely without affecting the original repository source.

Is this skill limited to image processing?

While designed for image processing pipelines, the pattern is universally applicable to any data-intensive scientific project requiring structured output directories.

How does this skill help with Git conflicts?

It ensures that large data files and modified notebooks are stored in user-specific directories outside the tracked source code, preventing binary file conflicts and accidental commits of sensitive data.

Does it support Windows environments?

Yes, it includes specific considerations for Windows, such as automated CUDA bin path detection and the use of pathlib for cross-platform directory resolution.

Scientific Project Data Separation

Name: Scientific Project Data Separation
Author: smith6jt-cop

bysmith6jt-cop

0•

Ciencia de Datos y ML

Organizes scientific research repositories by decoupling core code from experimental data and notebook outputs.

This skill implements a robust architectural pattern for scientific computing that strictly separates source code from user-generated experimental data, notebooks, and configurations. By using a centralized Project class and structured path management, it prevents Git conflicts, protects notebook templates from being overwritten, and ensures a standardized directory structure across diverse research environments. It is particularly effective for complex image processing pipelines where reproducibility and data isolation are critical, providing a canonical way to handle raw, processed, and analyzed data versions.

Características Principales

010 GitHub stars

02Centralized path management using Python's pathlib for platform-independent workflows

03Backward compatibility mapping for legacy notebook variables and directory names

04Template notebook copying to prevent accidental modification of core repository files

05JSON-based project configuration for persistent metadata and experiment tracking

06Automated directory structure generation for raw and processed data stages

Casos de Uso

01Managing large-scale scientific image processing pipelines like CODEX or microscopy

02Organizing collaborative research projects where code is shared via Git but data remains local

03Standardizing experimental outputs across different lab members or computing environments

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add smith6jt-cop/skills_registry project-data-separation

For use in Claude.ai and ChatGPT

Download Skill