n_data = 50
# true kappa
kappa_true = 0.3
# true noise added to data
sig_true = 2.0
# deterministic parameters
source_deterministic = 10.0
beta_deterministic = 2.0
bc_left_deterministic = 0.5
bc_right_deterministic = 0.0
# clean data
data_true = AdvDiff_solve(beta_deterministic, kappa_true, source_deterministic,
bc_left_deterministic, bc_right_deterministic)
# random noise
noise_true = np.random.normal(0, sig_true, n_data)
# noisy data
data_experiment = data_true + noise_true
def ln_prior(params):
# source : [-10, 10]
# sigma : [0, 10]
nrv = len(params0)
params = np.reshape(params, [1, nrv])
# print(params)
kappa = params[0, 0]
ln_prior_val = 0
def computational_model(beta, source, bc_left, bc_right, params):
return (AdvDiff_solve(beta, params[0, 0], source, bc_left, bc_right))
def computational_model(beta, kappa, bc_left, bc_right, params):
return (AdvDiff_solve(beta, kappa, params[0], bc_left, bc_right))