help with user specified probability distribution of galaxie
Posted: June 23 2010
I was wondering how to generate Cl's by specifying my own probability distribution of galaxies. (I see someone else has asked about this, but I haven't seen a response...) It looks like the code is already (somewhat) set up to do this, but I'm finding a few bugs. I don't know if someone has worked out the changes to make this run...?
To set things up I do the following:
; (1) set fiducial parameters:
fid=set_fiducial(cosmo_in={omega_m:.25d},calc_in={fit_tk:2},expt_in={sv1_n_zbin:2,sv1_zerror:0.02d})
; (2) create a survey structure:
sv=mk_survey(fid,'sv1',pz_input={z:zvals, pz:pzvals})
here zvals is a vector of z values and pzvals (for now) is a 2rowx401column matrix, (2 rows are the P(z) for each of two z bins and 401 is the length of the z vector):
This however crashes in mk_survey.pro at line 94:
"if keyword_set(pz_input) then d"
If I remove the "d" and combine line 94 with line 95 as:
"if keyword_set(pz_input) then pz=mk_pz(z,dndztype,dndzp,dndzz, n_g=n_g, z_med=z_med, z_min=z_min ,z_max=z_max ,pz_input=pz_input)"
then mk_pz will be called, but various errors keep it from running... I was wondering if someone might have debugged this portion of the code between 2008 (the date in the release) and now. Or is there another way of using this option?
To set things up I do the following:
; (1) set fiducial parameters:
fid=set_fiducial(cosmo_in={omega_m:.25d},calc_in={fit_tk:2},expt_in={sv1_n_zbin:2,sv1_zerror:0.02d})
; (2) create a survey structure:
sv=mk_survey(fid,'sv1',pz_input={z:zvals, pz:pzvals})
here zvals is a vector of z values and pzvals (for now) is a 2rowx401column matrix, (2 rows are the P(z) for each of two z bins and 401 is the length of the z vector):
This however crashes in mk_survey.pro at line 94:
"if keyword_set(pz_input) then d"
If I remove the "d" and combine line 94 with line 95 as:
"if keyword_set(pz_input) then pz=mk_pz(z,dndztype,dndzp,dndzz, n_g=n_g, z_med=z_med, z_min=z_min ,z_max=z_max ,pz_input=pz_input)"
then mk_pz will be called, but various errors keep it from running... I was wondering if someone might have debugged this portion of the code between 2008 (the date in the release) and now. Or is there another way of using this option?