Does this skill handle SDO sensor aging?

Yes, it includes degradation correction to compensate for sensor sensitivity changes over the mission lifetime using calibration tables.

How does registration improve ML model performance?

Registration ensures the solar disk is centered and oriented North-up, which eliminates spatial variance and helps neural networks focus on solar features rather than orientation artifacts.

Can I customize the output image resolution?

Yes, the pipeline supports various output resolutions such as 512, 1024, or 2048 pixels, with adjustable padding factors around the solar disk.

What are the primary dependencies for this skill?

The skill relies on the ssw-tools library, which integrates SunPy, Astropy, and aiapy for professional-grade solar data analysis.

What data formats does SSW-Prep support?

SSW-Prep primarily processes AIA Level 1 FITS data and outputs standardized FITS files optimized for machine learning workflows.

SSW Solar Data Preprocessing

Name: SSW Solar Data Preprocessing
Author: tykimos

bytykimos

0•

Data Science & ML

Transforms raw SDO/AIA solar observation data into standardized, ML-ready formats through automated calibration and registration pipelines.

SSW-Prep is a specialized tool for heliophysicists and data scientists using Claude Code to prepare SDO/AIA FITS data for machine learning workflows. It automates complex solar image processing tasks including pointing correction, north-up registration, sensor degradation compensation, and exposure normalization to ensure consistent inputs for neural networks. By streamlining the path from raw Level 1 data to high-quality ML datasets, it reduces the manual overhead of solar data science and ensures scientific accuracy in image analysis.

Key Features

01Batch processing support for large-scale solar datasets

02Sensor degradation and spacecraft pointing correction

03Solar disk registration and centering (North-up orientation)

04Exposure normalization from DN to DN/s

050 GitHub stars

06Automated AIA Level 1 FITS to ML-ready conversion

Use Cases

01Preparing training datasets for solar flare and space weather prediction models

02Calibrating archival solar data for long-term time-series analysis

03Standardizing multi-wavelength AIA images for convolutional neural networks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tykimos/ssw-plugin ssw-prep

For use in Claude.ai and ChatGPT

Download Skill