Muse tests on Log Normal simulator

[Justinezgh]

First test:

• notebook
• simulator with the following configuration:

  model = partial(
     lensingLogNormal, 
     N=128,  # number of latent variables
     map_size=5,
     gal_per_arcmin2=30,
     sigma_e=0.2,
     model_type='lognormal',
  ). 

[Justinezgh]

To do

• [ ] #2

• [ ] Check how muse behaves when we increase the amount of non-Gaussianity -> in Florent and Alan's paper, the gaussian field g is made log-Normal as follow:

  

  and they show that the larger alpha is, the more non-Guassianities there are.

  -> make the link between our 'shift' parameter and this 'amount of non-Gaussianity' parameter

  • [ ] #3