README.md

metadata

title: FOXES - A Framework for Operational X-ray Emission Synthesis
colorFrom: yellow
colorTo: red
sdk: gradio
sdk_version: 5.12.0
app_file: app.py
pinned: false
license: cc-by-4.0

FOXES - A Framework for Operational X-ray Emission Synthesis

Upload an SDO/AIA multi-wavelength observation to GOES soft X-ray (SXR) flux.

Model

• Architecture: ViT-based regression model
• Input: 7-channel AIA image (94, 131, 171, 193, 211, 304, 335 Angstrom) at 512x512 resolution
• Output: Predicted GOES soft X-ray flux (W/m^2)
• Patch size: 8x8 pixels (64x64 = 4096 patches)

Usage

Upload a .npy file containing a (7, 512, 512) float32 array of AIA wavelength images. The app will return:

1. Predicted SXR flux and GOES classification
2. AIA composite image (211/193/171 Angstrom as RGB)
3. Flux contribution map showing the spatial distribution of predicted flux

Setup

1. Upload normalized_sxr.npy to this Space
2. Upload your model checkpoint to a HF model repo (e.g., griffingoodwin04/FOXES-model) or place model.ckpt directly in this Space
3. (Optional) Add example .npy files to an examples/ folder