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, threadCount=1, init_pos=None, mpi_monch=False):
"""
runs mcmc on the parameter space given parameter bounds with CosmoHammerSampler
returns the chain
"""
lowerLimit, upperLimit = self.cosmoParam.param_bounds
params = np.array([mean_start, lowerLimit, upperLimit, sigma_start]).T
chain = LikelihoodComputationChain(
min=lowerLimit,
max=upperLimit)
temp_dir = tempfile.mkdtemp("Hammer")
file_prefix = os.path.join(temp_dir, "logs")
# chain.addCoreModule(CambCoreModule())
chain.addLikelihoodModule(self.chain)
chain.setup()
store = InMemoryStorageUtil()
if mpi_monch is True:
sampler = MpiCosmoHammerSampler(
params=params,
likelihoodComputationChain=chain,
filePrefix=file_prefix,
walkersRatio=walkerRatio,
burninIterations=n_burn,
sampleIterations=n_run,
threadCount=1,
initPositionGenerator=init_pos,
storageUtil=store)
else:
sampler = CosmoHammerSampler(
params=params,
likelihoodComputationChain=chain,
filePrefix=file_prefix,
walkersRatio=walkerRatio,
burninIterations=n_burn,
sampleIterations=n_run,
threadCount=threadCount,
initPositionGenerator=init_pos,
storageUtil=store)
time_start = time.time()
if sampler.isMaster():
print('Computing the MCMC...')
print('Number of walkers = ', len(mean_start)*walkerRatio)
print('Burn-in itterations: ', n_burn)
print('Sampling itterations:', n_run)
sampler.startSampling()
if sampler.isMaster():
time_end = time.time()
print(time_end - time_start, 'time taken for MCMC sampling')
# if sampler._sampler.pool is not None:
# sampler._sampler.pool.close()
try:
shutil.rmtree(temp_dir)
except Exception as ex:
print(ex)
pass
return store.samples
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
chain.params = params
chain.addLikelihoodModule(PseudoCmbModule())
chain.setup()
# find the best fit value and update our params knowledge
print("find best fit point")
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
def mcmc_CH(self,
walkerRatio,
n_run,
n_burn,
mean_start,
sigma_start,
threadCount=1,
init_pos=None,
mpi=False):
"""
runs mcmc on the parameter space given parameter bounds with CosmoHammerSampler
returns the chain
"""
lowerLimit, upperLimit = self.lower_limit, self.upper_limit
mean_start = np.maximum(lowerLimit, mean_start)
mean_start = np.minimum(upperLimit, mean_start)
low_start = mean_start - sigma_start
high_start = mean_start + sigma_start
low_start = np.maximum(lowerLimit, low_start)
high_start = np.minimum(upperLimit, high_start)
sigma_start = (high_start - low_start) / 2
mean_start = (high_start + low_start) / 2
params = np.array([mean_start, lowerLimit, upperLimit, sigma_start]).T
chain = LikelihoodComputationChain(min=lowerLimit, max=upperLimit)
temp_dir = tempfile.mkdtemp("Hammer")
file_prefix = os.path.join(temp_dir, "logs")
#file_prefix = "./lenstronomy_debug"
# chain.addCoreModule(CambCoreModule())
chain.addLikelihoodModule(self.chain)
chain.setup()
store = InMemoryStorageUtil()
#store = None
if mpi is True:
sampler = MpiCosmoHammerSampler(params=params,
likelihoodComputationChain=chain,
filePrefix=file_prefix,
walkersRatio=walkerRatio,
burninIterations=n_burn,
sampleIterations=n_run,
threadCount=1,
initPositionGenerator=init_pos,
storageUtil=store)
else:
sampler = CosmoHammerSampler(params=params,
likelihoodComputationChain=chain,
filePrefix=file_prefix,
walkersRatio=walkerRatio,
burninIterations=n_burn,
sampleIterations=n_run,
threadCount=threadCount,
initPositionGenerator=init_pos,
storageUtil=store)
time_start = time.time()
if sampler.isMaster():
print('Computing the MCMC...')
print('Number of walkers = ', len(mean_start) * walkerRatio)
print('Burn-in iterations: ', n_burn)
print('Sampling iterations:', n_run)
sampler.startSampling()
if sampler.isMaster():
time_end = time.time()
print(time_end - time_start, 'time taken for MCMC sampling')
# if sampler._sampler.pool is not None:
# sampler._sampler.pool.close()
try:
shutil.rmtree(temp_dir)
except Exception as ex:
print(ex, 'shutil.rmtree did not work')
pass
#samples = np.loadtxt(file_prefix+".out")
#prob = np.loadtxt(file_prefix+"prob.out")
return store.samples, store.prob
from cosmoHammer.util import Params
#parameter start center, min, max, start width
params = Params(("x", [1, -10, 10, 0.1]), ("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?")
chain.addLikelihoodModule(GSKYLike(saccfile_coadd, noise_saccfile_coadd))
chain.setup()
chaindir = os.path.join('chains', config['output_run_dir'])
if not os.path.isdir(get_output_fname(config, chaindir)):
os.makedirs(get_output_fname(config, chaindir))
path2chain = os.path.join('chains', config['output_run_dir'] + '/' + ch_config_params['chainsPrefix'])
if ch_config_params['use_mpi'] == 0:
if ch_config_params['rerun'] == 0:
sampler = CosmoHammerSampler(
params=params,
likelihoodComputationChain=chain,
filePrefix=get_output_fname(config, path2chain),
walkersRatio=ch_config_params['walkersRatio'],
burninIterations=ch_config_params['burninIterations'],
sampleIterations=ch_config_params['sampleIterations'])
else:
assert ch_config_params[
'path2rerunchain'] is not None, 'rerun is {}, but path to rerun chains not set. Aborting.'.format(
ch_config_params['rerun'])
path2chain = args.path2rerunchain
sampler = CosmoHammerSampler(
params=params,
likelihoodComputationChain=chain,
filePrefix=get_output_fname(config, path2chain),
walkersRatio=ch_config_params['walkersRatio'],
burninIterations=ch_config_params['burninIterations'],
("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?")
# ...and a positive definite, non-trivial covariance matrix.
cov = 0.5-np.random.rand(ndim**2).reshape((ndim, ndim))
cov = np.triu(cov)
cov += cov.T - np.diag(cov.diagonal())
cov = np.dot(cov,cov)
# Invert the covariance matrix first.
icov = np.linalg.inv(cov)
params = [[0, -5, 5, 1] ]*ndim
params = np.array(params)
chain = LikelihoodComputationChain(params[:,1],params[:,2])
multivarianteGaussian = MultivarianteGaussianModule(icov, means)
chain.addLikelihoodModule(multivarianteGaussian)
chain.setup()
sampler = CosmoHammerSampler(
params= params,
likelihoodComputationChain=chain,
filePrefix="multivariantGausian",
walkersRatio=50,
burninIterations=0,
sampleIterations=100,
)
print("start sampling")
sampler.startSampling()
print("done!")
logPosterior = LogPosteriorModule(data, threads=k)
# logPosterior = LogPosteriorModule(data,hypr)
chain.addCoreModule(coremodule)
chain.addLikelihoodModule(logPosterior)
chain.setup()
#######################################################################################
############################## SAMPLER ########################################
#-------- Writes Derived parameters into a file------
storageUtil = DerivedParamterFileUtil(samp)
#------------- Setup the sampler -----------
sampler = CHSampler(
params=params,
likelihoodComputationChain=chain,
filePrefix=samp,
walkersRatio=wr,
burninIterations=Nbi,
sampleIterations=Nit,
storageUtil=storageUtil,
initPositions=positions,
threadCount=1,
)
#----------------------- Start sampling-------------------------
print("start sampling")
t1 = time()
sampler.startSampling()
b = time() - t1
print("done!")
print 'Doing ', Nit, ' iterations takes ', b, ' seconds'
sys.exit(0)