This repo contains my introductory python textbook for astronomy students, which covers the basics of learning the language with an emphasis on astronomical applications.
Hello! I am learning how to use emcee and I came across your tutorial where you fit a single variable model to the data. I wanted to try out an example with multiple variables and so did this :
import numpy as np
x = np.linspace(0,10,100)
y = np.linspace(0,10,100)
z = np.linspace(0,10,100)
noise = np.random.rand(100)
f = 1.5*x + 3.0*y + 4.5*z + noise + 6
import sys
sys.path.append("PATH/TO/EMCEE")
import emcee
def model(params, var):
a, b, c, d = params
x, y, z = var
return a*x + b*y + c*z + d
def log_likelihood(params, var, f, ferr):
return ((f - model(params, var))/ferr)**2
def log_prior(params):
a, b, c, d = params
if (0 < a < 5) and (0 < b < 6) and (0 < c < 7) and (0 < d < 8):
return 0.0
else:
return -np.inf
def log_prob(params, var, f, ferr):
lp = log_prior(params)
if not np.isfinite(lp):
return -np.inf
else:
return lp + log_likelihood(params, var, f, ferr)
var = x, y, z
data = (var, f, noise)
nwalkers = 32
niter = 500
initial = np.array([1, 2, 3, 4])
ndim = len(initial)
p0 = [np.array(initial) + 1e-7 * np.random.randn(ndim) for i in range(nwalkers)]
def main(p0,nwalkers,niter,ndim,log_prob,data):
sampler = emcee.EnsembleSampler(nwalkers, ndim, log_prob, args=data)
print("Running burn-in...")
p0, _, _ = sampler.run_mcmc(p0, 100)
sampler.reset()
print("Running production...")
pos, prob, state = sampler.run_mcmc(p0, niter)
print("Run Complete!")
return sampler, pos, prob, state
sampler, pos, prob, state = main(p0,nwalkers,niter,ndim,log_prob,data)
and it is showing this error :
Running burn-in...
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in _get_lnprob(self, pos)
384 try:
--> 385 lnprob = np.array([float(l[0]) for l in results])
386 blob = [l[1] for l in results]
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in <listcomp>(.0)
384 try:
--> 385 lnprob = np.array([float(l[0]) for l in results])
386 blob = [l[1] for l in results]
TypeError: 'float' object is not subscriptable
During handling of the above exception, another exception occurred:
TypeError Traceback (most recent call last)
<ipython-input-73-9dd151c3debc> in <module>
58 return sampler, pos, prob, state
59
---> 60 sampler, pos, prob, state = main(p0,nwalkers,niter,ndim,log_prob,data)
<ipython-input-73-9dd151c3debc> in main(p0, nwalkers, niter, ndim, log_prob, data)
48
49 print("Running burn-in...")
---> 50 p0, _, _ = sampler.run_mcmc(p0, 100)
51 sampler.reset()
52
~/PATH/TO/lib/python3.9/site-packages/emcee/sampler.py in run_mcmc(self, pos0, N, rstate0, lnprob0, **kwargs)
169 rstate0 = self._last_run_mcmc_result[2]
170
--> 171 for results in self.sample(pos0, lnprob0, rstate0, iterations=N,
172 **kwargs):
173 pass
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in sample(self, p0, lnprob0, rstate0, blobs0, iterations, thin, storechain, mh_proposal)
256 first, second = slice(halfk), slice(halfk, self.k)
257 for S0, S1 in [(first, second), (second, first)]:
--> 258 q, newlnp, acc, blob = self._propose_stretch(p[S0], p[S1],
259 lnprob[S0])
260 if np.any(acc):
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in _propose_stretch(self, p0, p1, lnprob0)
330 # Calculate the proposed positions and the log-probability there.
331 q = c[rint] - zz[:, np.newaxis] * (c[rint] - s)
--> 332 newlnprob, blob = self._get_lnprob(q)
333
334 # Decide whether or not the proposals should be accepted.
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in _get_lnprob(self, pos)
386 blob = [l[1] for l in results]
387 except (IndexError, TypeError):
--> 388 lnprob = np.array([float(l) for l in results])
389 blob = None
390
~/PATH/TO/lib/python3.9/site-packages/emcee/ensemble.py in <listcomp>(.0)
386 blob = [l[1] for l in results]
387 except (IndexError, TypeError):
--> 388 lnprob = np.array([float(l) for l in results])
389 blob = None
390
TypeError: only size-1 arrays can be converted to Python scalars
I think I kinda understand the issue here - there is something wrong with the way I am passing the variables into the data holder (that goes into args). I was hoping you could help me understand what is going on and how to rectify it.
Hello! I am learning how to use
emceeand I came across your tutorial where you fit a single variable model to the data. I wanted to try out an example with multiple variables and so did this :and it is showing this error :
I think I kinda understand the issue here - there is something wrong with the way I am passing the variables into the data holder (that goes into args). I was hoping you could help me understand what is going on and how to rectify it.