estimate_list = [1] * n
# the range of penalization strength
lambda_range = [1e-8, 1e6]
# number of penalizations
n_lambda = 2
# initalize l1 regularization
# add shift parameter (=epsilon) which defines the offset of
# the logarithmic scale
initialize_regularization = regularization.l1_regularization(
model=model,
petab_problem=petab_problem,
objective=obj,
scale_list=scale_list_logicle,
estimate_list=estimate_list,
lambda_range=lambda_range,
n_lambda=n_lambda,
logicle_obj=logicle_obj)
# DO L1 REGULARIZATION _________________________________________________________________________________________________
# specify inputs
n_starts = 10
# optimization method and options
opt_method = 'L-BFGS-B'
opt_options = {'maxiter': 1e5, 'ftol': 1e-10, 'gtol': 1e-10, 'maxls': 80}
# opt.options = {'maxcor': 10, 'ftol': 1e-10, 'gtol': 1e-05, 'eps': 1e-08, 'maxfun': 1e5,
# 'maxiter': 1e5, 'maxls': 20}
estimate_list = [1, 1, 1, 0]
# the range of penalization strength
lambda_range = [1, 100]
# number of penalizations
n_lambda = 100
# initalize l1 regularization
# add shift parameter (=epsilon) which defines the offset of
# the logarithmic scale
initialize_regularization = regularization.l1_regularization(
model=model,
petab_problem=petab_problem,
objective=obj,
scale_list=scale_list_logE,
estimate_list=estimate_list,
lambda_range=lambda_range,
n_lambda=n_lambda,
shift_par=eps)
# DO L1 REGULARIZATION _________________________________________________________________________________________________
# specify inputs
n_starts = 1000
# optimization method and options
opt_method = 'L-BFGS-B'
opt_options = {'maxiter': 1e5, 'ftol': 1e-10, 'gtol': 1e-10}
# define startpoints as the best starts of the optimization