search

EoRImaging / eppsilon

eppsilon - error propagated power spectrum with interleaved observed noise
BSD 2-Clause "Simplified" License
5 stars 4 forks source link

Add mean back into kpar=0 for std_power #109

Closed bhazelton closed 4 years ago

bhazelton commented 5 years ago

Also drop uv pixels with less than 1/3 of the frequencies as we do in the Lomb-Scargle case.

closes #91

bhazelton commented 5 years ago

@nicholebarry can you look at this one too?

It appears that something is off in the noise ration for the kpar=0 bin, but that is true in master as well. Nothing jumps out at me in code as a mistake there, but I'll keep looking.

bhazelton commented 5 years ago

Nichole suggested replacing the standard power transform with just not rotating the cosine & sine modes to keep this code closer to the default Lomb-Scargle approach.

bhazelton commented 4 years ago

Closing this in favor of an approach suggested by @nicholebarry to keep the code more aligned: just don't rotate between sin & cos modes for standard power (i.e. set the rotation angle to zero).