search

BradGreig / Hybrid21CM

1 stars 3 forks source link

LF likelihood (prior) #13

Closed steven-murray closed 5 years ago

steven-murray commented 5 years ago

Need to implement this (most code is basically there).

steven-murray commented 5 years ago

@BradGreig when you have time, could you modify the computeLF function so that the return structure has both the density and the luminosity array? I feel like this is probably necessary for comparing to data.

BradGreig commented 5 years ago

Will do

BradGreig commented 5 years ago

Modified the ComputeLF to return M_UV, M_h and luminosity data for all redshifts passed to the function in commit 8f289cb4c49c3499e0c368fad44215776e6cee01. Passed as a 1D array, indexed by nbins*redshift. Indexing can be changed if required

steven-murray commented 5 years ago

OK so I think this is implemented/tested well enough as of 83343f98c1970c4d6ec4a85c24db11b0c3169ece. @BradGreig I'd be grateful if you could play around with it a little to make sure you think it's doing sensible things. In particular, at the moment the likelihood just accepts a keyword "sigma", which can be passed as a float - to give constant uncertainty per bin (in log-luminosity), or a function of Muv, or an array which has to match the length of Muv. I'm not sure if this is the most useful way of specifying things, or how you are used to doing it.

steven-murray commented 5 years ago

Nominally closing due to no activity.