andycasey · February 2, 2024 00:39
diff --git a/starfit_sampling.py b/starfit_sampling.py
 # Example to sample from the observed quantities and run each sample through StarFit

 # In this example we will say that we have the following abundances:
 #   [C/Fe] = -1.1 +/- 0.2
 #   [N/Fe] = -0.5 +/- 0.1
 #   [Mg/Fe] < 0.3

 # Whenever you're generating random numbers, you should always "seed" the random number generator.
 # That way you'll still get "random"-esque numbers, but if you run your code twice you will get the same "random" numbers.
 # If the numbers were really random each time you ran the code then it would be difficult to debug things.

 import numpy as np
 np.random.seed(314159) # pick your favourite magic number

 c_fe = np.random.normal(-1.1, 0.2)
 n_fe = np.random.normal(-0.5, 0.1)
 mg_fe = np.random.uniform(-12, 0.3) # here -12 is just some very unrealistically low value

 # Then you can repeat this process many times (~100) and see how much the StarFit results vary.
	# Example to sample from the observed quantities and run each sample through StarFit

	# In this example we will say that we have the following abundances:
	# [C/Fe] = -1.1 +/- 0.2
	# [N/Fe] = -0.5 +/- 0.1
	# [Mg/Fe] < 0.3

	# Whenever you're generating random numbers, you should always "seed" the random number generator.
	# That way you'll still get "random"-esque numbers, but if you run your code twice you will get the same "random" numbers.
	# If the numbers were really random each time you ran the code then it would be difficult to debug things.

	import numpy as np
	np.random.seed(314159) # pick your favourite magic number

	c_fe = np.random.normal(-1.1, 0.2)
	n_fe = np.random.normal(-0.5, 0.1)
	mg_fe = np.random.uniform(-12, 0.3) # here -12 is just some very unrealistically low value

	# Then you can repeat this process many times (~100) and see how much the StarFit results vary.