def test_grad_search(model, crit):
"""check that the paths are the same in the line search"""
n_outer = 2
criterion = HeldOutLogistic(X_val, y_val, model)
monitor1 = Monitor()
algo = Forward()
grad_search(algo, criterion, log_alpha, monitor1, n_outer=n_outer,
tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
monitor2 = Monitor()
algo = Implicit()
grad_search(algo, criterion, log_alpha, monitor2, n_outer=n_outer,
tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
monitor3 = Monitor()
algo = ImplicitForward(tol_jac=tol, n_iter_jac=5000)
grad_search(algo, criterion, log_alpha, monitor3, n_outer=n_outer,
tol=tol)
assert np.allclose(
np.array(monitor1.log_alphas), np.array(monitor3.log_alphas))
assert np.allclose(
np.array(monitor1.grads), np.array(monitor3.grads), atol=1e-4)
assert np.allclose(
np.array(monitor1.objs), np.array(monitor3.objs))
assert not np.allclose(
np.array(monitor1.times), np.array(monitor3.times))
def test_cross_val_criterion():
alpha_min = alpha_max / 10
log_alpha_max = np.log(alpha_max)
log_alpha_min = np.log(alpha_min)
max_iter = 10000
n_alphas = 10
kf = KFold(n_splits=5, shuffle=True, random_state=56)
estimator = sklearn.linear_model.Lasso(fit_intercept=False,
max_iter=1000,
warm_start=True)
monitor_grid = Monitor()
criterion = CrossVal(X, y, Lasso, cv=kf, estimator=estimator)
algo = Forward()
grid_search(algo,
criterion,
log_alpha_min,
log_alpha_max,
monitor_grid,
max_evals=n_alphas,
tol=tol)
reg = LassoCV(cv=kf,
verbose=True,
tol=tol,
fit_intercept=False,
alphas=np.geomspace(alpha_max, alpha_min, num=n_alphas),
max_iter=max_iter).fit(X, y)
reg.score(X, y)
objs_grid_sk = reg.mse_path_.mean(axis=1)
# these 2 value should be the same
(objs_grid_sk - np.array(monitor_grid.objs))
assert np.allclose(objs_grid_sk, monitor_grid.objs)
def test_grad_search(model):
# criterion = SURE(
# X_train, y_train, model, sigma=sigma_star, X_test=X_test,
# y_test=y_test)
n_outer = 3
criterion = HeldOutSmoothedHinge(X_val,
y_val,
model,
X_test=None,
y_test=None)
monitor1 = Monitor()
algo = Forward()
grad_search(algo,
criterion,
np.log(1e-3),
monitor1,
n_outer=n_outer,
tol=1e-13)
# criterion = SURE(
# X_train, y_train, model, sigma=sigma_star, X_test=X_test,
# y_test=y_test)
criterion = HeldOutSmoothedHinge(X_val,
y_val,
model,
X_test=None,
y_test=None)
monitor2 = Monitor()
algo = Implicit()
grad_search(algo,
criterion,
np.log(1e-3),
monitor2,
n_outer=n_outer,
tol=1e-13)
# criterion = SURE(
# X_train, y_train, model, sigma=sigma_star, X_test=X_test,
# y_test=y_test)
criterion = HeldOutSmoothedHinge(X_val,
y_val,
model,
X_test=None,
y_test=None)
monitor3 = Monitor()
algo = ImplicitForward(tol_jac=1e-6, n_iter_jac=100)
grad_search(algo,
criterion,
np.log(1e-3),
monitor3,
n_outer=n_outer,
tol=1e-13)
assert np.allclose(np.array(monitor1.log_alphas),
np.array(monitor3.log_alphas))
assert np.allclose(np.array(monitor1.grads), np.array(monitor3.grads))
assert np.allclose(np.array(monitor1.objs), np.array(monitor3.objs))
# assert np.allclose(
# np.array(monitor1.objs_test), np.array(monitor3.objs_test))
assert not np.allclose(np.array(monitor1.times), np.array(monitor3.times))
def test_val_grad(model):
#######################################################################
# Not all methods computes the full Jacobian, but all
# compute the gradients
# check that the gradient returned by all methods are the same
criterion = HeldOutLogistic(X_val, y_val, model)
algo = Forward()
val_fwd, grad_fwd = criterion.get_val_grad(log_C,
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
algo = ImplicitForward(tol_jac=1e-8, n_iter_jac=100)
val_imp_fwd, grad_imp_fwd = criterion.get_val_grad(log_C,
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
algo = Implicit()
val_imp, grad_imp = criterion.get_val_grad(log_C,
algo.get_beta_jac_v,
tol=tol)
assert np.allclose(val_fwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_imp_fwd)
assert np.allclose(val_imp_fwd, val_imp)
assert np.allclose(grad_imp_fwd, grad_imp, atol=1e-5)
def test_grad_search(model, crit):
"""check that the paths are the same in the line search"""
if crit == 'cv':
n_outer = 2
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
else:
n_outer = 2
criterion = SURE(X_train,
y_train,
model,
sigma=sigma_star,
X_test=X_test,
y_test=y_test)
criterion = HeldOutMSE(X_val, y_val, model, X_test=X_test, y_test=y_test)
monitor1 = Monitor()
algo = Forward()
grad_search(algo,
criterion,
log_alpha,
monitor1,
n_outer=n_outer,
tol=1e-16)
criterion = HeldOutMSE(X_val, y_val, model, X_test=X_test, y_test=y_test)
monitor2 = Monitor()
algo = Implicit()
grad_search(algo,
criterion,
log_alpha,
monitor2,
n_outer=n_outer,
tol=1e-16)
criterion = HeldOutMSE(X_val, y_val, model, X_test=X_test, y_test=y_test)
monitor3 = Monitor()
algo = ImplicitForward(tol_jac=1e-8, n_iter_jac=5000)
grad_search(algo,
criterion,
log_alpha,
monitor3,
n_outer=n_outer,
tol=1e-16)
assert np.allclose(np.array(monitor1.log_alphas),
np.array(monitor3.log_alphas))
assert np.allclose(np.array(monitor1.grads), np.array(monitor3.grads))
assert np.allclose(np.array(monitor1.objs), np.array(monitor3.objs))
assert np.allclose(np.array(monitor1.objs_test),
np.array(monitor3.objs_test))
assert not np.allclose(np.array(monitor1.times), np.array(monitor3.times))
def test_grad_search():
n_outer = 3
criterion = HeldOutMSE(X_val, y_val, model, X_test=None, y_test=None)
monitor1 = Monitor()
algo = Forward()
grad_search(algo,
criterion,
np.array([log_alpha1, log_alpha2]),
monitor1,
n_outer=n_outer,
tol=1e-16)
criterion = HeldOutMSE(X_val, y_val, model, X_test=None, y_test=None)
monitor2 = Monitor()
algo = Implicit()
grad_search(algo,
criterion,
np.array([log_alpha1, log_alpha2]),
monitor2,
n_outer=n_outer,
tol=1e-16)
criterion = HeldOutMSE(X_val, y_val, model, X_test=None, y_test=None)
monitor3 = Monitor()
algo = ImplicitForward(tol_jac=1e-3, n_iter_jac=1000)
grad_search(algo,
criterion,
np.array([log_alpha1, log_alpha2]),
monitor3,
n_outer=n_outer,
tol=1e-16)
[np.linalg.norm(grad) for grad in monitor1.grads]
[np.exp(alpha) for alpha in monitor1.log_alphas]
assert np.allclose(np.array(monitor1.log_alphas),
np.array(monitor3.log_alphas))
assert np.allclose(np.array(monitor1.grads), np.array(monitor3.grads))
assert np.allclose(np.array(monitor1.objs), np.array(monitor3.objs))
assert not np.allclose(np.array(monitor1.times), np.array(monitor3.times))
assert np.allclose(np.array(monitor1.log_alphas),
np.array(monitor2.log_alphas),
atol=1e-2)
assert np.allclose(np.array(monitor1.grads),
np.array(monitor2.grads),
atol=1e-2)
assert np.allclose(np.array(monitor1.objs),
np.array(monitor2.objs),
atol=1e-2)
assert not np.allclose(np.array(monitor1.times), np.array(monitor2.times))
def test_val_grad():
#######################################################################
# Not all methods computes the full Jacobian, but all
# compute the gradients
# check that the gradient returned by all methods are the same
criterion = HeldOutMSE(X_val, y_val, model)
algo = Forward()
val_fwd, grad_fwd = criterion.get_val_grad(np.array(
[log_alpha1, log_alpha2]),
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutMSE(X_val, y_val, model)
algo = ImplicitForward(tol_jac=1e-16, n_iter_jac=5000)
val_imp_fwd, grad_imp_fwd = criterion.get_val_grad(np.array(
[log_alpha1, log_alpha2]),
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutMSE(X_val, y_val, model)
algo = ImplicitForward(tol_jac=1e-16, n_iter_jac=5000)
val_imp_fwd_custom, grad_imp_fwd_custom = criterion.get_val_grad(
np.array([log_alpha1, log_alpha2]), algo.get_beta_jac_v, tol=tol)
criterion = HeldOutMSE(X_val, y_val, model)
algo = Implicit()
val_imp, grad_imp = criterion.get_val_grad(np.array(
[log_alpha1, log_alpha2]),
algo.get_beta_jac_v,
tol=tol)
assert np.allclose(val_fwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_imp_fwd)
assert np.allclose(val_imp_fwd, val_imp)
assert np.allclose(val_imp_fwd, val_imp_fwd_custom)
# for the implcit the conjugate grad does not converge
# hence the rtol=1e-2
assert np.allclose(grad_imp_fwd, grad_imp, atol=1e-3)
assert np.allclose(grad_imp_fwd, grad_imp_fwd_custom)
def test_val_grad(model):
criterion = HeldOutLogistic(X_val, y_val, model)
algo = Forward()
val_fwd, grad_fwd = criterion.get_val_grad(
log_alpha, algo.get_beta_jac_v, tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
algo = ImplicitForward(tol_jac=1e-8, n_iter_jac=5000)
val_imp_fwd, grad_imp_fwd = criterion.get_val_grad(
log_alpha, algo.get_beta_jac_v, tol=tol)
criterion = HeldOutLogistic(X_val, y_val, model)
algo = Implicit()
val_imp, grad_imp = criterion.get_val_grad(
log_alpha, algo.get_beta_jac_v, tol=tol)
assert np.allclose(val_fwd, val_imp_fwd, atol=1e-4)
assert np.allclose(grad_fwd, grad_imp_fwd, atol=1e-4)
assert np.allclose(val_imp_fwd, val_imp, atol=1e-4)
# for the implcit the conjugate grad does not converge
# hence the rtol=1e-2
assert np.allclose(grad_imp_fwd, grad_imp, rtol=1e-2)
log_alphas = np.log(alphas)
##############################################################################
# Grid-search
# -----------
print('scikit started')
t0 = time.time()
estimator = LogisticRegression(penalty='l1',
fit_intercept=False,
solver='saga',
max_iter=max_iter)
model = SparseLogreg(X_train, y_train, max_iter=max_iter, estimator=estimator)
criterion = HeldOutLogistic(X_val, y_val, model)
algo_grid = Forward()
monitor_grid = Monitor()
grid_search(algo_grid,
criterion,
log_alpha_min,
log_alpha_max,
monitor_grid,
log_alphas=log_alphas,
tol=tol)
objs = np.array(monitor_grid.objs)
t_sk = time.time() - t0
print('scikit finished')
print("Time to compute CV for scikit-learn: %.2f" % t_sk)
def test_val_grad():
#######################################################################
# Not all methods computes the full Jacobian, but all
# compute the gradients
# check that the gradient returned by all methods are the same
for key in models.keys():
# model = Lasso(log_alpha)
log_alpha = dict_log_alpha[key]
model = models[key]
# model = Lasso(log_alpha)
criterion = HeldOutMSE(X_val, y_val, model)
algo = Forward()
val_fwd, grad_fwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutMSE(X_val, y_val, model)
algo = ImplicitForward(tol_jac=1e-8, n_iter_jac=5000)
val_imp_fwd, grad_imp_fwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
criterion = HeldOutMSE(X_val, y_val, model)
algo = Implicit()
val_imp, grad_imp = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
# import ipdb; ipdb.set_trace()
criterion = HeldOutMSE(X_val, y_val, model)
algo = Backward()
val_bwd, grad_bwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
assert np.allclose(val_fwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_imp_fwd)
# assert np.allclose(val_imp_fwd, val_imp)
assert np.allclose(val_bwd, val_fwd)
assert np.allclose(val_bwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_bwd)
assert np.allclose(grad_bwd, grad_imp_fwd)
# for the implcit the conjugate grad does not converge
# hence the rtol=1e-2
assert np.allclose(grad_imp_fwd, grad_imp, atol=1e-3)
for key in models.keys():
# model = Lasso(log_alpha)
log_alpha = dict_log_alpha[key]
model = models[key]
# model = Lasso(log_alpha)
criterion = SURE(X_train, y_train, model, sigma_star)
algo = Forward()
val_fwd, grad_fwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
criterion = SURE(X_train, y_train, model, sigma_star)
algo = ImplicitForward(tol_jac=1e-8, n_iter_jac=5000)
val_imp_fwd, grad_imp_fwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
criterion = SURE(X_train, y_train, model, sigma_star)
algo = Implicit(criterion)
val_imp, grad_imp = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
criterion = SURE(X_train, y_train, model, sigma_star)
algo = Backward()
val_bwd, grad_bwd = criterion.get_val_grad(log_alpha,
algo.get_beta_jac_v,
tol=tol)
assert np.allclose(val_fwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_imp_fwd)
assert np.allclose(val_imp_fwd, val_imp)
assert np.allclose(val_bwd, val_fwd)
assert np.allclose(val_bwd, val_imp_fwd)
assert np.allclose(grad_fwd, grad_bwd)
assert np.allclose(grad_bwd, grad_imp_fwd)
def parallel_function(dataset_name,
method,
tol=1e-5,
n_outer=50,
tolerance_decrease='exponential'):
# load data
X_train, X_val, X_test, y_train, y_val, y_test = get_data(dataset_name)
n_samples, n_features = X_train.shape
print('n_samples', n_samples)
print('n_features', n_features)
y_train[y_train == 0.0] = -1.0
y_val[y_val == 0.0] = -1.0
y_test[y_test == 0.0] = -1.0
alpha_max = np.max(np.abs(X_train.T @ y_train))
alpha_max /= X_train.shape[0]
log_alpha_max = np.log(alpha_max)
alpha_min = alpha_max * 1e-2
# alphas = np.geomspace(alpha_max, alpha_min, 10)
# log_alphas = np.log(alphas)
log_alpha1_0 = np.log(0.1 * alpha_max)
log_alpha2_0 = np.log(0.1 * alpha_max)
log_alpha_max = np.log(alpha_max)
n_outer = 25
if dataset_name == "rcv1":
size_loop = 2
else:
size_loop = 2
model = ElasticNet(X_train,
y_train,
log_alpha1_0,
log_alpha2_0,
log_alpha_max,
max_iter=1000,
tol=tol)
for i in range(size_loop):
monitor = Monitor()
if method == "implicit_forward":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = ImplicitForward(criterion, tol_jac=1e-3, n_iter_jac=100)
_, _, _ = grad_search(algo=algo,
verbose=False,
log_alpha0=np.array(
[log_alpha1_0, log_alpha2_0]),
tol=tol,
n_outer=n_outer,
monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "forward":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = Forward(criterion)
_, _, _ = grad_search(algo=algo,
log_alpha0=np.array(
[log_alpha1_0, log_alpha2_0]),
tol=tol,
n_outer=n_outer,
monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "implicit":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = Implicit(criterion)
_, _, _ = grad_search(algo=algo,
log_alpha0=np.array(
[log_alpha1_0, log_alpha2_0]),
tol=tol,
n_outer=n_outer,
monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "grid_search":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(alpha_min)
log_alpha_opt, min_g_func = grid_search(
algo,
log_alpha_min,
log_alpha_max,
monitor,
max_evals=10,
tol=tol,
samp="grid",
t_max=dict_t_max[dataset_name],
log_alphas=None,
nb_hyperparam=2)
print(log_alpha_opt)
elif method == "random":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(alpha_min)
log_alpha_opt, min_g_func = grid_search(
algo,
log_alpha_min,
np.log(alpha_max),
monitor,
max_evals=10,
tol=tol,
samp="random",
t_max=dict_t_max[dataset_name],
nb_hyperparam=2)
print(log_alpha_opt)
elif method == "lhs":
criterion = HeldOutMSE(X_val,
y_val,
model,
X_test=X_test,
y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(alpha_min)
log_alpha_opt, min_g_func = grid_search(
algo,
log_alpha_min,
np.log(alpha_max),
monitor,
max_evals=10,
tol=tol,
samp="lhs",
t_max=dict_t_max[dataset_name])
print(log_alpha_opt)
monitor.times = np.array(monitor.times).copy()
monitor.objs = np.array(monitor.objs).copy()
monitor.objs_test = np.array(monitor.objs_test).copy()
monitor.log_alphas = np.array(monitor.log_alphas).copy()
return (dataset_name, method, tol, n_outer, tolerance_decrease,
monitor.times, monitor.objs, monitor.objs_test, monitor.log_alphas,
norm(y_val), norm(y_test), log_alpha_max)
# X_train, X_val, X_test, y_train, y_val, y_test = get_leukemia()
n_samples, n_features = X_train.shape
print("Starting path computation...")
alpha_max = np.max(np.abs(X_train.T @ (-y_train)))
alpha_max /= (2 * n_samples)
n_alphas = 10
p_alphas = np.geomspace(1, 1e-4, n_alphas)
alphas = p_alphas * alpha_max
log_alphas = np.log(alphas)
tol = 1e-5
# grid search
model = SparseLogreg(X_train, y_train, log_alphas[0], max_iter=1000)
criterion = HeldOutLogistic(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward(criterion)
monitor_grid_sk = Monitor()
grid_search(algo, None, None, monitor_grid_sk, log_alphas=log_alphas, tol=tol)
monitor = Monitor()
# grad_search(
# algo, logCs[0], monitor, n_outer=5, verbose=True,
# tolerance_decrease='constant', tol=1e-8,
# t_max=10000)
plt.figure()
plt.plot(monitor_grid_sk.log_alphas, monitor_grid_sk.objs)
plt.plot(monitor.log_alphas, monitor.objs, 'bo')
plt.show(block=False)
def test_grid_search():
max_evals = 5
monitor_grid = Monitor()
model = Lasso(X_train, y_train, estimator=estimator)
criterion = HeldOutMSE(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward()
log_alpha_opt_grid, _ = grid_search(algo,
criterion,
log_alpha_min,
log_alpha_max,
monitor_grid,
max_evals=max_evals,
tol=1e-5,
samp="grid")
monitor_random = Monitor()
criterion = HeldOutMSE(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward()
log_alpha_opt_random, _ = grid_search(algo,
criterion,
log_alpha_min,
log_alpha_max,
monitor_random,
max_evals=max_evals,
tol=1e-5,
samp="random")
assert (monitor_random.log_alphas[np.argmin(
monitor_random.objs)] == log_alpha_opt_random)
assert (monitor_grid.log_alphas[np.argmin(
monitor_grid.objs)] == log_alpha_opt_grid)
monitor_grid = Monitor()
model = Lasso(X_train, y_train, estimator=estimator)
criterion = SURE(X_train,
y_train,
model,
sigma=sigma_star,
X_test=X_test,
y_test=y_test)
algo = Forward()
log_alpha_opt_grid, _ = grid_search(algo,
criterion,
log_alpha_min,
log_alpha_max,
monitor_grid,
max_evals=max_evals,
tol=1e-5,
samp="grid")
monitor_random = Monitor()
criterion = SURE(X_train,
y_train,
model,
sigma=sigma_star,
X_test=X_test,
y_test=y_test)
algo = Forward()
log_alpha_opt_random, _ = grid_search(algo,
criterion,
log_alpha_min,
log_alpha_max,
monitor_random,
max_evals=max_evals,
tol=1e-5,
samp="random")
assert (monitor_random.log_alphas[np.argmin(
monitor_random.objs)] == log_alpha_opt_random)
assert (monitor_grid.log_alphas[np.argmin(
monitor_grid.objs)] == log_alpha_opt_grid)
def parallel_function(
dataset_name, method, tol=1e-5, n_outer=50,
tolerance_decrease='exponential'):
# load data
X_train, X_val, X_test, y_train, y_val, y_test = get_data(dataset_name, csr=True)
n_samples, n_features = X_train.shape
print('n_samples', n_samples)
print('n_features', n_features)
y_train[y_train == 0.0] = -1.0
y_val[y_val == 0.0] = -1.0
y_test[y_test == 0.0] = -1.0
C_max = 100
logC = np.log(1e-2)
n_outer = 5
if dataset_name == "rcv1":
size_loop = 1
else:
size_loop = 1
model = SVM(
X_train, y_train, logC, max_iter=10000, tol=tol)
for i in range(size_loop):
monitor = Monitor()
if method == "implicit_forward":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = ImplicitForward(criterion, tol_jac=1e-3, n_iter_jac=100)
_, _, _ = grad_search(
algo=algo, verbose=False,
log_alpha0=logC, tol=tol,
n_outer=n_outer, monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "forward":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward(criterion)
_, _, _ = grad_search(
algo=algo,
log_alpha0=logC, tol=tol,
n_outer=n_outer, monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "implicit":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Implicit(criterion)
_, _, _ = grad_search(
algo=algo,
log_alpha0=logC, tol=tol,
n_outer=n_outer, monitor=monitor,
t_max=dict_t_max[dataset_name],
tolerance_decrease=tolerance_decrease)
elif method == "grid_search":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(1e-2)
log_alpha_opt, min_g_func = grid_search(
algo, log_alpha_min, np.log(C_max), monitor, max_evals=25,
tol=tol, samp="grid")
print(log_alpha_opt)
elif method == "random":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(1e-2)
log_alpha_opt, min_g_func = grid_search(
algo, log_alpha_min, np.log(C_max), monitor, max_evals=25,
tol=tol, samp="random")
print(log_alpha_opt)
elif method == "lhs":
criterion = HeldOutSmoothedHinge(X_val, y_val, model, X_test=X_test, y_test=y_test)
algo = Forward(criterion)
log_alpha_min = np.log(1e-2)
log_alpha_opt, min_g_func = grid_search(
algo, log_alpha_min, np.log(C_max), monitor, max_evals=25,
tol=tol, samp="lhs")
print(log_alpha_opt)
monitor.times = np.array(monitor.times)
monitor.objs = np.array(monitor.objs)
monitor.objs_test = np.array(monitor.objs_test)
monitor.log_alphas = np.array(monitor.log_alphas)
return (dataset_name, method, tol, n_outer, tolerance_decrease,
monitor.times, monitor.objs, monitor.objs_test,
monitor.log_alphas, norm(y_val), norm(y_test))