Once #5 is resolved, test the next step of emulator training.
My intuition is that because we are now dealing with a narrow range in wavelength, any given chunk should be explained by relatively few eigenspectra. Hopefully this bears out in practice.
Second, there will likely need to be some configuration changes internal to Starfish (mostly in the direction of simplifying things) since we have modified config.yaml.
Once #5 is resolved, test the next step of emulator training.
My intuition is that because we are now dealing with a narrow range in wavelength, any given chunk should be explained by relatively few eigenspectra. Hopefully this bears out in practice.
Second, there will likely need to be some configuration changes internal to Starfish (mostly in the direction of simplifying things) since we have modified
config.yaml.