ssw-plugin:ssw-prep

Preprocess SDO/AIA Level 1 FITS files into ML-ready float32 format.

Updated Feb 6, 2026

Search Tags:#machine learning #calibration #preprocessing #FITS #solar data #SDO/AIA

Ready to run

npx skills add https://github.com/tykimos/ssw-plugin --skill ssw-plugin-ssw-prep

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ssw-plugin:ssw-prep
Source: https://github.com/tykimos/ssw-plugin/tree/main/skills/ssw-prep
Command: npx skills add https://github.com/tykimos/ssw-plugin --skill ssw-plugin-ssw-prep

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sunpy, astropy, aiapy, ssw-tools, loguru, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the complex preprocessing of SDO/AIA solar data, transforming raw FITS files into a standardized, machine-learning-ready format.

Core Features & Use Cases

Calibration: Corrects for pointing errors, sensor degradation, and exposure variations.
Normalization: Standardizes images to a consistent resolution, orientation (north-up), and solar disk centering.
Format Conversion: Converts raw FITS data into float32 format with units of DN/s, suitable for neural network training.
Use Case: Prepare a batch of SDO/AIA 193 Angstrom images taken over a week for training a solar flare prediction model.

Quick Start

Use the ssw-prep skill to preprocess the AIA Level 1 FITS file 'aia_lev1_file.fits' into an ML-ready format.

Frequently Asked Questions about ssw-plugin:ssw-prep

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I preprocess SDO/AIA FITS files for machine learning training?▼

Preprocess SDO/AIA FITS files by calibrating pointing, degradation, and exposure, then normalizing the solar disk to a standardized resolution and orientation for ML training.

What does solar data calibration involve for SDO/AIA Level 1 images?▼

Solar data calibration corrects Level 1 FITS images for pointing errors, sensor degradation, and exposure variations to ensure consistent input data for machine learning models.

How do I convert raw FITS solar data into a float32 format for neural networks?▼

Convert raw FITS solar data into float32 format with units of DN/s by applying calibration, registering the solar disk, and normalizing the images to a consistent resolution and north-up orientation.

Can I batch convert SDO/AIA FITS files for solar flare prediction models?▼

Yes, you can batch convert SDO/AIA FITS files by calibrating and normalizing multiple raw images into a standardized ML-ready format suitable for training solar flare prediction models.

Do I need sunpy and astropy to prepare SDO/AIA data for ML?▼

Yes, preparing SDO/AIA data requires sunpy, astropy, and aiapy to handle FITS file processing, perform solar physics calibrations, and register the solar disk for machine learning workflows.