def main():
parser = argparse.ArgumentParser()
parser.add_argument('propNO', nargs='?', type=int, default=1)
parser.add_argument('num_samp', nargs='?', type=int, default=10000)
parser.add_argument('num_burnin', nargs='?', type=int, default=10000)
parser.add_argument('step_sizes', nargs='?', type=float, default=[.1,.2]) # SNR10: [.5,1.]; SNR100: [.1,.2]
# parser.add_argument('step_nums', nargs='?', type=int, default=[1])
parser.add_argument('props', nargs='?', type=str, default=['LI_prior', 'LI_Langevin'])
args = parser.parse_args()
## define the inverse elliptic problem ##
# parameters for PDE model
nx=40;ny=40;
# parameters for prior model
sigma=1.25;s=0.0625
# parameters for misfit model
SNR=100 # 100
# define the inverse problem
elliptic=Elliptic(nx=nx,ny=ny,SNR=SNR,sigma=sigma,s=s)
# initialization
# unknown=elliptic.prior.sample(whiten=True)
unknown=elliptic.prior.gen_vector()
# unknown=df.Function(elliptic.pde.V)
# MAP_file=os.path.join(os.getcwd(),'result/MAP_SNR'+str(SNR)+'.h5')
# if os.path.isfile(MAP_file):
# f=df.HDF5File(elliptic.pde.mpi_comm,MAP_file,"r")
# f.read(unknown,'parameter')
# f.close()
# else:
# unknown=elliptic.get_MAP(SAVE=True)
# unknown=elliptic.prior.u2v(unknown.vector())
# run MCMC to generate samples
print("Preparing DILI sampler using {} proposal with step sizes {} for LIS and CS resp....".format(args.props[args.propNO],args.step_sizes,))
dili=DILI(unknown,elliptic,args.step_sizes,proposal=args.props[args.propNO],n_lag=100,n_max=100)
dili.adaptive_MCMC(args.num_samp,args.num_burnin,threshold_l=1e-3,threshold_g=1e-3)
# append PDE information including the count of solving
filename_=os.path.join(dili.savepath,dili.filename+'.pckl')
filename=os.path.join(dili.savepath,'Elliptic_'+dili.filename+'.pckl') # change filename
os.rename(filename_, filename)
f=open(filename,'ab')
soln_count=elliptic.pde.soln_count
pickle.dump([nx,ny,sigma,s,SNR,soln_count,args],f)
f.close()
Shiwei Lan @ CalTech, 2017
"""
import os,pickle
import dolfin as df
import numpy as np
import matplotlib.pyplot as plt
# import pandas as pd
from itertools import cycle
from Elliptic_dili import Elliptic
# define the inverse problem
np.random.seed(2017)
SNR=100
elliptic = Elliptic(nx=40,ny=40,SNR=SNR)
# algorithms
algs=('pCN','infMALA','infHMC','DRinfmMALA','DRinfmHMC','DILI','aDRinfmMALA','aDRinfmHMC')
alg_names=('pCN','$\infty$-MALA','$\infty$-HMC','DR-$\infty$-mMALA','DR-$\infty$-mHMC','DILI','aDR-$\infty$-mMALA','aDR-$\infty$-mHMC')
num_algs=len(algs)
found = np.zeros(num_algs,dtype=np.bool)
# preparation for estimates
folder = './analysis_f_SNR'+str(SNR)
h5_names=[f for f in os.listdir(folder) if f.endswith('.h5')]
num_samp=2000
pckl_names=[f for f in os.listdir(folder) if f.endswith('.pckl')]
max_iter=1000
max_time=1500
def main():
parser = argparse.ArgumentParser()
parser.add_argument('algNO', nargs='?', type=int, default=0)
parser.add_argument('propNO', nargs='?', type=int, default=1)
parser.add_argument('num_samp', nargs='?', type=int, default=10000)
parser.add_argument('num_burnin', nargs='?', type=int, default=10000)
parser.add_argument('step_sizes', nargs='?', type=float,
default=[.25,
.1]) # SNR10: [3.,1.5]; SNR100: [.22,.06]
parser.add_argument('step_nums', nargs='?', type=int, default=[1, 4])
parser.add_argument('algs',
nargs='?',
type=str,
default=('aDRinfmMALA', 'aDRinfmHMC'))
parser.add_argument('props',
nargs='?',
type=str,
default=('LI_prior', 'LI_Langevin'))
args = parser.parse_args()
## define the inverse elliptic problem ##
# parameters for PDE model
nx = 40
ny = 40
# parameters for prior model
sigma = 1.25
s = 0.0625
# parameters for misfit model
SNR = 100 # 100
# define the inverse problem
elliptic = Elliptic(nx=nx, ny=ny, SNR=SNR, sigma=sigma, s=s)
# initialization
# unknown=elliptic.prior.sample(whiten=True)
unknown = elliptic.prior.gen_vector()
# unknown=df.Function(elliptic.pde.V)
# MAP_file=os.path.join(os.getcwd(),'result/MAP_SNR'+str(SNR)+'.h5')
# if os.path.isfile(MAP_file):
# f=df.HDF5File(elliptic.pde.mpi_comm,MAP_file,"r")
# f.read(unknown,'parameter')
# f.close()
# else:
# unknown=elliptic.get_MAP(SAVE=True)
# unknown=elliptic.prior.u2v(unknown.vector())
# run MCMC to generate samples
print(
"Preparing %s sampler using %s proposal with step size %g for %d step(s)..."
% (args.algs[args.algNO], args.props[args.propNO],
args.step_sizes[args.algNO], args.step_nums[args.algNO]))
# if args.algs[args.algNO] is 'aDRinfmMALA':
# adrinfmGMC=aDRinfmMALA(unknown,elliptic,args.step_size,proposal=args.props[args.propNO])
# elif args.algs[args.algNO] is 'aDRinfmHMC':
# adrinfmGMC=aDRinfmHMC(unknown,elliptic,args.step_size,args.step_nums[args.algNO],proposal=args.props[args.propNO])
# else:
# print('Algorithm not available!')
# raise
adrinfmGMC = aDRinfGMC(unknown,
elliptic,
args.step_sizes[args.algNO],
args.step_nums[args.algNO],
args.algs[args.algNO],
args.props[args.propNO],
n_lag=100,
n_max=100)
adrinfmGMC.adaptive_MCMC(args.num_samp,
args.num_burnin,
threshold_l=1e-3,
threshold_g=1e-3)
# append PDE information including the count of solving
filename_ = os.path.join(adrinfmGMC.savepath,
adrinfmGMC.filename + '.pckl')
filename = os.path.join(adrinfmGMC.savepath, 'Elliptic_' +
adrinfmGMC.filename + '.pckl') # change filename
os.rename(filename_, filename)
f = open(filename, 'ab')
soln_count = elliptic.pde.soln_count
pickle.dump([nx, ny, sigma, s, SNR, soln_count, args], f)
f.close()
def main():
parser = argparse.ArgumentParser()
parser.add_argument('algNO', nargs='?', type=int, default=0)
parser.add_argument('num_samp', nargs='?', type=int, default=2000)
parser.add_argument('num_burnin', nargs='?', type=int, default=500)
parser.add_argument(
'step_sizes', nargs='?', type=float,
default=[.5, 2., 1.3, 6.,
4.]) # SNR10: [.5,2,1.3,6.,4.];SNR100: [.01,.04,0.04,.52,.25]
parser.add_argument('step_nums',
nargs='?',
type=int,
default=[1, 1, 4, 1, 4, 1, 4])
parser.add_argument('algs',
nargs='?',
type=str,
default=('pCN', 'infMALA', 'infHMC', 'DRinfmMALA',
'DRinfmHMC'))
args = parser.parse_args()
## define the inverse elliptic problem ##
# parameters for PDE model
nx = 40
ny = 40
# parameters for prior model
sigma = 1.25
s = 0.0625
# parameters for misfit model
SNR = 10 # 100
# define the inverse problem
elliptic = Elliptic(nx=nx, ny=ny, SNR=SNR, sigma=sigma, s=s)
# initialization
# unknown=elliptic.prior.sample(whiten=False)
unknown = elliptic.prior.gen_vector()
# unknown=df.Function(elliptic.pde.V)
# MAP_file=os.path.join(os.getcwd(),'result/MAP.h5')
# if os.path.isfile(MAP_file):
# f=df.HDF5File(elliptic.pde.mpi_comm,MAP_file,"r")
# f.read(unknown,'parameter')
# f.close()
# else:
# unknown=elliptic.get_MAP(SAVE=True)
# run MCMC to generate samples
print("Preparing %s sampler with step size %g for %d step(s)..." %
(args.algs[args.algNO], args.step_sizes[args.algNO],
args.step_nums[args.algNO]))
inf_GMC = geoinfMC(unknown,
elliptic,
args.step_sizes[args.algNO],
args.step_nums[args.algNO],
args.algs[args.algNO],
k=5)
mc_fun = inf_GMC.sample
mc_args = (args.num_samp, args.num_burnin)
mc_fun(*mc_args)
# append PDE information including the count of solving
filename_ = os.path.join(inf_GMC.savepath, inf_GMC.filename + '.pckl')
filename = os.path.join(inf_GMC.savepath, 'Elliptic_' + inf_GMC.filename +
'.pckl') # change filename
os.rename(filename_, filename)
f = open(filename, 'ab')
# soln_count=[elliptic.soln_count,elliptic.pde.soln_count]
soln_count = elliptic.pde.soln_count
pickle.dump([nx, ny, sigma, s, SNR, soln_count, args], f)
f.close()
"""
eigs=kwargs.pop('eigs',None)
if eigs:# is not None:
return _GA_posterior_lr(prior,eigs,**kwargs)
metact=kwargs.pop('metact',None)
rtmetact=kwargs.pop('rtmetact',None)
if metact and rtmetact:
return _GA_posterior_exct(prior,metact,rtmetact,**kwargs)
else:
df.error('Definition not specified!')
if __name__ == '__main__':
# np.random.seed(2017)
from Elliptic_dili import Elliptic
# define the inverse problem
elliptic=Elliptic(nx=40,ny=40,SNR=10)
# get MAP
unknown=df.Function(elliptic.pde.V)
MAP_file=os.path.join(os.getcwd(),'MAP.h5')
if os.path.isfile(MAP_file):
f=df.HDF5File(elliptic.pde.mpi_comm,MAP_file,"r")
f.read(unknown,'parameter')
f.close()
else:
unknown=elliptic.get_MAP(SAVE=True)
import time
start = time.time()
# get eigen-decomposition of posterior Hessian at MAP
_,_,_,eigs=elliptic.get_geom(unknown.vector(),geom_ord=[1.5],whitened=False,k=100)#threshold=1e-2)
# define approximate Gaussian posterior