def specie_h(population, p, n, pset, direccion): # heuristic applied by specie
gpo_specie=specie_gpo(population)
for specie in gpo_specie:
eval_prob(specie[0])
for ind in population:
if ind.bestspecie_get()==1:
if random.random()<=ind.get_LS_prob():
strg=ind.__str__()
args=[]
if len(pset.arguments) > 0:
for arg in pset.arguments:
args.append(arg)
l_strg=add_subt_cf(strg, args)
c = tree2f()
cd=c.convert(l_strg)
xdata,ydata=get_address(p, n,direccion)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_,cd , xdata, ydata, p0=ind.get_params(), method='trf', max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
ind.params_set(beta_opt)
funcEval.cont_evalp += nfev
def specie_h(population, p, n, pset, direccion): # heuristic applied by specie
gpo_specie=specie_gpo(population)
for specie in gpo_specie:
eval_prob(specie[0])
for ind in population:
if ind.bestspecie_get()==1:
if random.random()<=ind.get_LS_prob():
strg=ind.__str__()
args=[]
if len(pset.arguments) > 0:
for arg in pset.arguments:
args.append(arg)
l_strg=add_subt_cf(strg, args)
c = tree2f()
cd=c.convert(l_strg)
xdata,ydata=get_address(p, n,direccion)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_,cd , xdata, ydata, p0=ind.get_params(), method='trf', max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
ind.params_set(beta_opt)
funcEval.cont_evalp += nfev
def ls_random(population, p, n, pset, direccion, problem,
benchmark_flag): # size heuristic
random_set = random_set_pop(population)
for ind in random_set:
strg = ind.__str__()
args = []
if len(pset.arguments) > 1:
for arg in pset.arguments:
args.append(arg)
l_strg = add_subt_cf(strg, args)
c = tree2f()
cd = c.convert(l_strg)
xdata, ydata = get_address(p, n, problem, direccion, benchmark_flag)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_,
cd,
xdata,
ydata,
p0=ind.get_params(),
method='trf',
max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
funcEval.cont_evalp += nfev
ind.params_set(beta_opt)
def best_specie(population, p, n, pset, direccion, problem,
benchmark_flag): # best of each specie
for ind in population:
if ind.bestspecie_get() == 1:
strg = ind.__str__()
args = []
if len(pset.arguments) > 0:
for arg in pset.arguments:
args.append(arg)
l_strg = add_subt_cf(strg, args)
c = tree2f()
cd = c.convert(l_strg)
xdata, ydata = get_address(p, n, problem, direccion,
benchmark_flag)
beta_opt, beta_cov, success, nfev = curve_fit_2(
eval_,
cd,
xdata,
ydata,
p0=ind.get_params(),
method='trf',
max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
ind.params_set(beta_opt)
funcEval.cont_evalp += nfev
def best_pop_ls(population, p, n, pset, direccion): # best of the pop
ind = best_pop(population)
strg = ind.__str__()
args = []
if len(pset.arguments) > 1:
for arg in pset.arguments:
args.append(arg)
l_strg = add_subt_cf(strg, args)
c = tree2f()
cd = c.convert(l_strg)
xdata,ydata = get_address(p, n, direccion)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_, cd, xdata, ydata, p0=ind.get_params(), method='trf', max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
ind.params_set(beta_opt)
funcEval.cont_evalp += nfev
def best_pop_ls(population, p, n, pset, direccion): # best of the pop
ind = best_pop(population)
strg = ind.__str__()
args = []
if len(pset.arguments) > 1:
for arg in pset.arguments:
args.append(arg)
l_strg = add_subt_cf(strg, args)
c = tree2f()
cd = c.convert(l_strg)
xdata,ydata = get_address(p, n, direccion)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_, cd, xdata, ydata, p0=ind.get_params(), method='trf', max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
ind.params_set(beta_opt)
funcEval.cont_evalp += nfev
def ls_random(population, p, n, pset, direccion, problem): # size heuristic
random_set=random_set_pop(population)
for ind in random_set:
strg = ind.__str__()
args = []
if len(pset.arguments) > 1:
for arg in pset.arguments:
args.append(arg)
l_strg = add_subt_cf(strg, args)
c = tree2f()
cd = c.convert(l_strg)
xdata, ydata = get_address(p, n, problem,direccion)
beta_opt, beta_cov, success, nfev = curve_fit_2(eval_,cd , xdata, ydata, p0=ind.get_params(), method='trf', max_nfev=40)
if not success:
ind.LS_applied_set(1)
ind.LS_story_set(1)
else:
ind.LS_applied_set(1)
ind.LS_story_set(1)
funcEval.cont_evalp += nfev
ind.params_set(beta_opt)
