Train CHARM for ltu-cmass

maho3 / ltu-gobig

1 stars 0 forks source link

Train CHARM for ltu-cmass #3

Closed maho3 closed 7 months ago

maho3 commented 8 months ago

CHARM is a flow-based halo-biasing model. We want to train it on 8 Mpc/h resolution, with input dark matter fields rom the ltu-cmass simulators and output halo fields from the quijote HR latin hypercube

Definition of done:

Plots showing the 1-pt and 2-pt statistics of halos generated by CHARM and drawn from Quijote

maho3 commented 7 months ago

This has been done by @shivampcosmo

This is being integrated into ltu-cmass in https://github.com/maho3/ltu-cmass/issues/9