def sampler(self, walker_ratio, burnin, samples, num, threads=-1):
"""
:param walker_ratio: the ratio of walkers and the count of sampled parameters
:param burnin: burin iterations
:param samples: no. of sample iterations
:param num: number to put in output files e.g: string(name+num)=Pk_1,Bk_1
:param threads: no. of cpu threads
self.chain.setup()
print("find best fit point")
pso = MpiParticleSwarmOptimizer(self.chain, params[:, 1], params[:, 2])
psoTrace = np.array([pso.gbest.position.copy() for _ in pso.sample()])
params[:, 0] = pso.gbest.position
"""
sampler = CosmoHammerSampler(params=params,
likelihoodComputationChain=self.chain,
filePrefix='%s' % self.name + '%d' % num,
walkersRatio=walker_ratio,
burninIterations=burnin,
sampleIterations=samples,
threadCount=threads)
print("started sampling:")
start = time.time()
sampler.startSampling()
end = time.time()
tics = end - start
print("The time taken %.2f sec. done!" % tics)
print('Done!')
def mcmc_CH(self,
walkerRatio,
n_run,
n_burn,
mean_start,
sigma_start,
lowerLimit,
upperLimit,
X2_chain,
threadCount=1,
init_pos=None):
"""
runs mcmc on the parameter space given parameter bounds with CosmoHammerSampler
returns the chain
"""
params = np.array([mean_start, lowerLimit, upperLimit, sigma_start]).T
chain = LikelihoodComputationChain(min=lowerLimit, max=upperLimit)
# chain.addCoreModule(CambCoreModule())
chain.addLikelihoodModule(X2_chain)
chain.setup()
store = InMemoryStorageUtil()
sampler = CosmoHammerSampler(params=params,
likelihoodComputationChain=chain,
filePrefix="testHammer",
walkersRatio=walkerRatio,
burninIterations=n_burn,
sampleIterations=n_run,
threadCount=threadCount,
initPositionGenerator=init_pos,
storageUtil=store)
sampler.startSampling()
if sampler._sampler.pool is not None:
sampler._sampler.pool.close()
return store.samples
pso = ParticleSwarmOptimizer(chain, params[:, 1], params[:, 2])
psoTrace = np.array([pso.gbest.position.copy() for _ in pso.sample()])
params[:, 0] = pso.gbest.position
storageUtil = InMemoryStorageUtil()
sampler = CosmoHammerSampler(params=params,
likelihoodComputationChain=chain,
filePrefix="pseudoCmb_pso",
walkersRatio=50,
burninIterations=0,
sampleIterations=100,
storageUtil=storageUtil,
threadCount=4)
print("start sampling")
sampler.startSampling()
print("done!")
print("plotting")
import matplotlib.pyplot as plt
import triangle
data = storageUtil.samples
K = data.shape[1]
factor = 2.0 # size of one side of one panel
lbdim = 0.5 * factor # size of left/bottom margin
trdim = 0.2 * factor # size of top/right margin
whspace = 0.05 # w/hspace size
plotdim = factor * K + factor * (K - 1.) * whspace
dim = lbdim + plotdim + trdim
fig, ax = plt.subplots(K, K, figsize=(dim, dim), tight_layout=False)
("y", [1, -10, 10, 0.1]))
chain = LikelihoodComputationChain()
rosenbrock = RosenbrockModule()
chain.addLikelihoodModule(rosenbrock)
chain.setup()
sampler = CosmoHammerSampler(
params= params,
likelihoodComputationChain=chain,
filePrefix="rosenbrock",
walkersRatio=50,
burninIterations=100,
sampleIterations=100)
sampler.startSampling()
try:
import numpy as np
import matplotlib.pyplot as plt
data = np.loadtxt("rosenbrock.out")
plt.scatter(data[:,0], data[:, 1], 1)
plt.xlabel(params.keys[0])
plt.ylabel(params.keys[1])
plt.show()
except Exception:
print("Plotting failed. Missing libs?")