m1_4kandClaude Sonnet 4.6
a46d9b3796
- k-space (Fourier-domain) random subsampling (10% default) - ISTA reconstruction with TV (Chambolle) regularisation - Sampling patterns: variable_density (MRI-style) and uniform - Adjoint operator verified to machine epsilon - scikit-image TV denoiser with pure-NumPy Chambolle fallback Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
6 lines
89 B
Plaintext
6 lines
89 B
Plaintext
# Required
|
|
numpy
|
|
|
|
# Strongly recommended (better/faster TV denoising)
|
|
scikit-image>=0.19
|