| CI | status |
|---|---|
| pip builds |  |
Python bindings for LIRA with SIMD intrinsics.
sudo apt install libtbb-dev r-base-dev r-base r-mathlib
git clone --recursive https://github.com/infinitron/pylira_simd
pip install ./pylira_simd--TBU--
from pylira_simd import image_analysis, LiraPayload, get_test_payload
import numpy as np
##############################################################
# usage 1 (recommended)
# Create a LiraPayload object by supplying at least an input image,
# a baseline image, a psf, and output file names. The images
# can be either fits file names or numpy arrays
...
payload = LiraPayload(inp_image,baseline_image,psf_image,\
out_im_file,out_param)
# Displays input images and prints a table with input params
payload.describe_payload()
payload.launch_image_analysis()
###############################################################
# For PSF needing to to be updated dynamically
payload = LiraPayload(inp_image,baseline_image,None,\
out_im_file,out_param)
payload.use_prag_bayesian_psf = True
psf_arr = ...
# define a function that returns a new PSF every 10 iterations
def prg_bayes_psf(i):
if i%10 == 0: #i=0 must always return a non-empty PSF
return psf_arr[i%10] #can be from an array, external variable or function
else:
return np.ndarray((0,0))
payload.prag_bayes_psf_func = prg_bayes_psf
payload.launch_image_analysis()
###############################################################
# usage 2
# image_analysis is the low level function used by LiraPayload
# to run LIRA on the inputs.
...
...
image_analysis(...)
# Test the program with sample data
payload = get_test_payload()
payload.launch_image_analysis()