model,
X,
y,
None,
None,
monitor,
max_evals=20,
tol=tol,
alphas=alphas)
elif algo_name == 'random_search' or algo_name == 'bayesian':
hyperopt_wrapper(algo,
criterion,
model,
X,
y,
alpha_min,
alpha_max,
monitor,
max_evals=20,
tol=tol,
method=algo_name,
size_space=2)
else:
1 / 0
dict_monitor[algo_name] = monitor
min_objs = np.infty
for monitor in dict_monitor.values():
monitor.objs = np.array(monitor.objs)
min_objs = min(min_objs, monitor.objs.min())
scaling_factor = (y @ y) / len(y)
n_alphas = 100
p_alphas = np.geomspace(1, p_alpha_min, n_alphas)
alphas = alpha_max * p_alphas
reg = LassoCV(cv=kf,
verbose=True,
tol=tol,
fit_intercept=False,
alphas=alphas,
max_iter=max_iter).fit(X, y)
reg.score(X, y)
objs = reg.mse_path_.mean(axis=1)
log_alphas = np.log(alphas)
else:
hyperopt_wrapper(algo,
cross_val_criterion,
model,
X,
y,
np.log(alpha_max * p_alpha_min),
np.log(alpha_max),
monitor,
max_evals=10,
method=algorithm,
size_space=1)
objs = np.array(monitor.objs)
log_alphas = np.array(monitor.log_alphas)
log_alphas -= np.log(alpha_max)
np.save("results/log_alphas_%s" % algorithm, log_alphas)
np.save("results/objs_%s" % algorithm, objs)
print('%s finished' % algorithm)
optimizer = LineSearch(n_outer=25, verbose=True, tol=tol)
grad_search(
algo, cross_val_criterion, model, optimizer, X, y, log_alpha0,
monitor)
elif algorithm.startswith('grid_search'):
if algorithm == 'grid_search10':
n_alphas = 5
else:
n_alphas = 30
p_alphas = np.geomspace(1, p_alpha_min, n_alphas)
alphas = alpha_max * p_alphas
log_alphas = np.log(alphas)
grid_alphas = [i for i in itertools.product(log_alphas, log_alphas)]
grid_search(
algo, cross_val_criterion, model, X, y, None, None, monitor,
log_alphas=grid_alphas)
else:
hyperopt_wrapper(
algo, cross_val_criterion, model, X, y, log_alpha_min,
log_alpha_max, monitor, max_evals=max_evals,
method=algorithm, size_space=2)
objs = np.array(monitor.objs)
log_alphas = np.array(monitor.log_alphas)
log_alphas -= np.log(alpha_max)
np.save("results/%s_log_alphas_%s_enet" % (dataset, algorithm), log_alphas)
np.save("results/%s_objs_%s_enet" % (dataset, algorithm), objs)
print('%s finished' % algorithm)
criterion,
model,
X,
y,
alpha_min,
alpha_max,
monitor,
max_evals=20,
tol=tol)
elif algo_name == 'random_search':
hyperopt_wrapper(algo,
criterion,
model,
X,
y,
alpha_min,
alpha_max,
monitor,
max_evals=20,
tol=tol,
method='random',
size_space=1)
dict_monitor[algo_name] = monitor
min_objs = np.infty
for monitor in dict_monitor.values():
monitor.objs = np.array(monitor.objs)
min_objs = min(min_objs, monitor.objs.min())
scaling_factor = (y @ y) / len(y)
plt.figure()
for monitor in dict_monitor.values():
p_alphas = np.geomspace(1, p_alpha_min, n_alphas)
alphas = alpha_max * p_alphas
reg = LassoCV(cv=kf,
verbose=True,
tol=tol,
fit_intercept=False,
alphas=alphas,
max_iter=max_iter).fit(X, y)
reg.score(X, y)
objs = reg.mse_path_.mean(axis=1)
log_alphas = np.log(alphas / alpha_max)
else:
hyperopt_wrapper(algo,
cross_val_criterion,
model,
X,
y,
alpha_max * p_alpha_min,
alpha_max,
monitor,
max_evals=10,
method=algorithm,
size_space=1,
tol=tol,
random_state=4)
objs = np.array(monitor.objs)
log_alphas = np.log(np.array(monitor.alphas) / alpha_max)
np.save("results/log_alphas_%s" % algorithm, log_alphas)
np.save("results/objs_%s" % algorithm, objs)
print('%s finished' % algorithm)