Probability distribution functions

The PDF folder contains the estimated probability distribution functions (PDFs). The data is organized following the general hierarchical structure described in Suite folders.

We provide PDFs for all the 3D grid files in the CAMELS Multifield Dataset (CMD) (See CAMELS Multifield Dataset), each containing 1,000 grids from the LH set.

The PDFs are stored as .npy files and they are named as hist_Grids_prefix_sim_LH_grid_z=redshift.npy, where prefix is the word identifying each field (HI, Vgas, etc.), sim can be IllustrisTNG, SIMBA, Nbody_IllustrisTNG, or Nbody_SIMBA, grid can be 128, 256, or 512 and redshift can be 0, 0.5, 1, 1.5 or 2. These files can be read with numpy as follows:

import numpy as np

# name of the pdf file
pdfgrids = 'hist_Grids_HI_SIMBA_LH_128_z=0.00.npy'

# read the data
pdf = np.load(pdfgrids) #--shape of pdf is (1000,500)

# get the bin counts for the 4th grid
print(pdf[3])

The file contains 1,000 entries with 500 values per entry denoting the number of counts in 500 bins for that entry. See the CAMELS data release for more details on how the PDFs are calculated.