Skip to content

Commit

Permalink
update readme
Browse files Browse the repository at this point in the history
  • Loading branch information
thibautlouis committed Mar 12, 2024
1 parent 1750599 commit ee602a7
Showing 1 changed file with 14 additions and 8 deletions.
22 changes: 14 additions & 8 deletions project/data_analysis/dr6xplanck.rst
Original file line number Diff line number Diff line change
Expand Up @@ -43,8 +43,6 @@ ACT/Planck cross correlation
OMP_NUM_THREADS=48 srun -n 5 -c 48 --cpu-bind=cores python get_window_dr6.py global_dr6v4xlegacy.dict
# real 10m2.348s
OMP_NUM_THREADS=32 srun -n 8 -c 32 --cpu-bind=cores python get_mcm_and_bbl.py global_dr6v4xlegacy.dict
.. code:: shell
salloc --nodes 1 --qos interactive --time 02:00:00 --constraint cpu
OMP_NUM_THREADS=20 srun -n 12 -c 20 --cpu-bind=cores python get_mcm_and_bbl.py global_dr6v4xlegacy.dict
Expand Down Expand Up @@ -73,6 +71,8 @@ ACT/Planck cross correlation
salloc --nodes 4 --qos interactive --time 03:00:00 --constraint cpu
OMP_NUM_THREADS=32 srun -n 32 -c 32 --cpu-bind=cores python get_covariance_blocks.py global_dr6v4xlegacy.dict
# real 13m24.803s
to correct for the leakage, grab the code in the leakage folder

.. code:: shell
Expand All @@ -84,22 +84,31 @@ ACT/Planck cross correlation
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu-bind=cores python get_leakage_covariance.py global_dr6v4xlegacy.dict
# real 18m12.066s
the planck spectra can have leftover systematic in them, we have estimated this using planck end-to-end simuation (see bottom of this page), grab the code in the planck folder

.. code:: shell
salloc --nodes 1 --qos interactive --time 01:00:00 --constraint cpu
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu_bind=cores python get_corrected_planck_spectra.py global_dr6v4xlegacy.dict
Now to calibrate and get the expected polarisation efficiencies, grab the code in the calibration folder

.. code:: shell
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu_bind=cores python get_calibs.py global_dr6v4xlegacy.dict
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu_bind=cores python get_polar_eff_LCDM.py global_dr6v4xlegacy.dict
In addition to the standard dr6 simulation tools (e.g:)
.. code:: shell
salloc --nodes 2 --qos interactive --time 4:00:00 --constraint cpu
OMP_NUM_THREADS=128 srun -n 4 -c 128 --cpu_bind=cores python mc_mnms_get_nlms.py global_dr6_v4.dict
# real time ~ 3h (for 100 sims)
This directory also provides tools to get noise alms from NPIPE/legacy simulations to get montecarlo corrected errors. This script writes the alms in the same format as the ``mnms`` noise alms such that we can directly use the standard simulation script to get Planck simulations.
we have code to get planks simulation nlms

.. code:: shell
salloc -N 1 -C cpu -q interactive -t 03:00:00
Expand All @@ -109,7 +118,7 @@ This directory also provides tools to get noise alms from NPIPE/legacy simulatio
*******************
Xtra correction
End-to-end sim correction
*******************


Expand All @@ -124,8 +133,5 @@ the ACT spectra, the way we get the correction is the following
salloc -N 1 -C cpu -q interactive -t 01:00:00
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu_bind=cores python mc_analysis.py global_dr6v4xnpipe.dict
The first code is similar to the standard simulation spectra code, but it's residual only (no signal), the mc_analysis serve to produce the average of these spectra, then to correct the planck spectra run
The first code is similar to the standard simulation spectra code, but it's residual only (no signal), the mc_analysis serve to produce the average of these spectra.

.. code:: shell
salloc -N 1 -C cpu -q interactive -t 01:00:00
OMP_NUM_THREADS=256 srun -n 1 -c 256 --cpu_bind=cores python get_corrected_planck_spectra.py global_dr6v4xnpipe.dict

0 comments on commit ee602a7

Please sign in to comment.