cortecs: A Python package for compressing opacities

A Python package for decreasing the memory footprint of opacity functions. The primary functionality is compressing opacity functions with varying flexibility. Current methods include

• polynomial fitting
• PCA-based fitting
• neural network fitting

All fits are currently made in along the temperature and pressure axes.

Additionally, cortecs can chunk up opacity functions. The radiative transfer problem can often be cast as embarassingly parallel, so each chunk can be sent to a different CPU.