Create a photon list for a single pointing¶

In order to create the photon list, I interpolate spectra in temperature and abundance. Please refer to the corresponding notebook for the creation of the table of spectra used for interpolation

Imports¶

In [1]:

import matplotlib.pyplot as plt
import matplotlib.colors as colors
from astropy.io import fits
import jax.numpy as jnp
import numpy as np
import cmasher as cmr
import jax.random as random

from xifu_cluster_sim.xifu_config import XIFU_Config
from xifu_cluster_sim.cluster_cube import ClusterCube
from xifu_cluster_sim.photon_list import PhotonList, create_photon_list

import matplotlib.pyplot as plt import matplotlib.colors as colors from astropy.io import fits import jax.numpy as jnp import numpy as np import cmasher as cmr import jax.random as random from xifu_cluster_sim.xifu_config import XIFU_Config from xifu_cluster_sim.cluster_cube import ClusterCube from xifu_cluster_sim.photon_list import PhotonList, create_photon_list

Load (or create) cluster object¶

Loading¶

In [2]:

cluster = ClusterCube()
cluster.load_cube('path_to_cube.npy')

cluster = ClusterCube() cluster.load_cube('path_to_cube.npy')

---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
Cell In[2], line 2
      1 cluster = ClusterCube()
----> 2 cluster.load_cube('path_to_cube.npy')

File /xifu/home/mola/xifu_cluster_sim/src/xifu_cluster_sim/cluster_cube.py:276, in ClusterCube.load_cube(self, path)
    270 def load_cube(self, path):
    271     """
    272     Loads the cubes of computed value of the cluster
    273 
    274     """
--> 276     data  = np.load(path)
    277     self.squared_density_cube = data['ne_nH_cube']
    278     self.norm = data['norm_cube']

File ~/miniconda3/envs/myenv/lib/python3.10/site-packages/numpy/lib/npyio.py:427, in load(file, mmap_mode, allow_pickle, fix_imports, encoding, max_header_size)
    425     own_fid = False
    426 else:
--> 427     fid = stack.enter_context(open(os_fspath(file), "rb"))
    428     own_fid = True
    430 # Code to distinguish from NumPy binary files and pickles.

FileNotFoundError: [Errno 2] No such file or directory: 'path_to_cube.npy'

Or creating¶

In [7]:

cluster = ClusterCube()
cluster.create_density_cube()
cluster.create_norm_cube()
cluster.create_kT_cube()
cluster.create_Z_cube()
rng_key = random.PRNGKey(42)
cluster.create_velocity_cube(rng_key)
cluster.convert_velocity_to_redshift()

cluster = ClusterCube() cluster.create_density_cube() cluster.create_norm_cube() cluster.create_kT_cube() cluster.create_Z_cube() rng_key = random.PRNGKey(42) cluster.create_velocity_cube(rng_key) cluster.convert_velocity_to_redshift()

Create photon list¶

Specify the energy grid before, this grid shoulf be the same as the one used for the interpolated table of spectra. X-IFU range of energy is 0.2 to 12 keV. The energy step needs to be much smaller than the energy resolution.

In [8]:

E_low = 0.2
E_high = 12.
E_step = 4e-4
E_n_steps = int((E_high-E_low)/E_step + 1)
E_bounds = np.linspace(E_low,
                       E_high, 
                       E_n_steps) 

E_low = 0.2 E_high = 12. E_step = 4e-4 E_n_steps = int((E_high-E_low)/E_step + 1) E_bounds = np.linspace(E_low, E_high, E_n_steps)

Load the table of spectra for interpolation

In [9]:

data = np.load('../../data/spectres_interpoles_T_Z.npz')

T_scale = data['T_scale']
Z_scale = data['Z_scale']
spectra = data['spectra']

data = np.load('../../data/spectres_interpoles_T_Z.npz') T_scale = data['T_scale'] Z_scale = data['Z_scale'] spectra = data['spectra']

Initialize the photon list class

In [10]:

ph_list = PhotonList(cluster,
                    E_bounds,
                    Z_scale,
                    T_scale, 
                    spectra)

ph_list = PhotonList(cluster, E_bounds, Z_scale, T_scale, spectra)

Choose a pointing (the cluster cube can be larger than a single X-IFU pointing, to allow for several pointings to be simulated.

In [11]:

ph_list.select_sub_pointing(x_offset=0,
                            y_offset=0,
                            pointing_shape = (58,58)
                           )

ph_list.select_sub_pointing(x_offset=0, y_offset=0, pointing_shape = (58,58) )

Create the photon list. For memory management purposes, the actual function to call for that is outside the PhotonList class. The architecture used (multiprocessor + multithreaded) is not compatible with progress bars unfortunately. It should not take more than a minute in the following example on 32 cores.

In [12]:

# Call the arguments for parallelization 
#(this is essentially a division of the list of spectral properties to use for the model on several processors)
args = ph_list.get_args_list(numprocs = 8, 
                             numthreads = 4)
ph_list = create_photon_list(ph_list,
                            args, 
                            numproc = 8)

# Call the arguments for parallelization #(this is essentially a division of the list of spectral properties to use for the model on several processors) args = ph_list.get_args_list(numprocs = 8, numthreads = 4) ph_list = create_photon_list(ph_list, args, numproc = 8)

/home/mola/miniconda3/envs/myenv/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
  self.pid = os.fork()
/home/mola/miniconda3/envs/myenv/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
  self.pid = os.fork()

Write the photon list to simputfiles that can be used by SIXTE. Since I use multiprocessing throughout all this project, I divide the photon list into N_procs parts.

In [14]:

ph_list.write_photon_list_to_simputs(results,
                                     num_divisions=32,
                                     path = sim_path,
                                     name_format = 'ph_list')

ph_list.write_photon_list_to_simputs(results, num_divisions=32, path = sim_path, name_format = 'ph_list')

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[14], line 1
----> 1 ph_list.write_photon_list_to_simputs(results,
      2                                      num_divisions=32,
      3                                      path = sim_path,
      4                                      name_format = 'ph_list')

AttributeError: 'numpy.ndarray' object has no attribute 'write_photon_list_to_simputs'

In [ ]: