def test_warm_start_l2r():
clf = PrimalLinearSVC(warm_start=True, random_state=0, penalty="l2")
clf.C = 0.1
clf.fit(bin_dense, bin_target)
assert_almost_equal(clf.score(bin_dense, bin_target), 1.0)
clf.C = 0.2
clf.fit(bin_dense, bin_target)
assert_almost_equal(clf.score(bin_dense, bin_target), 1.0)
def test_fit_rbf_binary_l1r():
clf = PrimalSVC(C=0.5, kernel="rbf", gamma=0.1, random_state=0, penalty="l1")
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 0.845)
n_nz = np.sum(clf.coef_ != 0)
assert_equal(n_nz, 160)
K = pairwise_kernels(bin_dense, metric="rbf", gamma=0.1)
clf2 = PrimalLinearSVC(C=0.5, random_state=0, penalty="l1")
clf2.fit(K, bin_target)
acc = clf2.score(K, bin_target)
assert_almost_equal(acc, 0.845)
n_nz = np.sum(clf2.coef_ != 0)
assert_equal(n_nz, 160)
def test_fit_linear_binary_l2r():
clf = PrimalLinearSVC(C=1.0, random_state=0, penalty="l2")
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 1.0)
K = pairwise_kernels(bin_dense, metric="rbf", gamma=0.1)
clf2 = PrimalLinearSVC(C=0.5, random_state=0, penalty="l2")
clf2.fit(K, bin_target)
acc = clf2.score(K, bin_target)
assert_almost_equal(acc, 1.0)
n_nz = np.sum(clf2.coef_ != 0)
assert_equal(n_nz, 200)
def test_fit_linear_binary_l1r():
clf = PrimalLinearSVC(C=1.0, random_state=0, penalty="l1")
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 1.0)
n_nz = np.sum(clf.coef_ != 0)
clf = PrimalLinearSVC(C=0.1, random_state=0, penalty="l1")
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 0.97)
n_nz2 = np.sum(clf.coef_ != 0)
assert_true(n_nz > n_nz2)
def test_warm_start_l1r():
clf = PrimalLinearSVC(warm_start=True, random_state=0, penalty="l1")
clf.C = 0.1
clf.fit(bin_dense, bin_target)
n_nz = np.sum(clf.coef_ != 0)
clf.C = 0.2
clf.fit(bin_dense, bin_target)
n_nz2 = np.sum(clf.coef_ != 0)
assert_true(n_nz < n_nz2)
def test_lower_bound_binary():
Cmin = C_lower_bound(bin_dense, bin_target)
clf = PrimalLinearSVC(C=Cmin, random_state=0, penalty="l1")
clf.fit(bin_dense, bin_target)
n_nz = np.sum(clf.coef_ != 0)
assert_equal(0, n_nz)
clf = PrimalLinearSVC(C=Cmin * 2, random_state=0, penalty="l1")
clf.fit(bin_dense, bin_target)
n_nz = np.sum(clf.coef_ != 0)
assert_not_equal(0, n_nz)
def test_fit_squared_loss():
clf = PrimalLinearSVC(C=1.0, random_state=0, penalty="l2",
loss="squared", max_iter=100)
clf.fit(bin_dense, bin_target)
assert_almost_equal(clf.score(bin_dense, bin_target), 0.99)
y = bin_target.copy()
y[y == 0] = -1
assert_array_almost_equal(1 - y * np.dot(bin_dense, clf.coef_.ravel()),
clf.errors_.ravel())
K = pairwise_kernels(bin_dense, metric="rbf", gamma=0.1)
for kernel_regularizer in (True, False):
clf = PrimalSVC(C=1.0, random_state=0, penalty="l2",
kernel="rbf", gamma=0.1,
selection="loss",
loss="squared", max_iter=100,
kernel_regularizer=kernel_regularizer)
clf.fit(bin_dense, bin_target)
assert_almost_equal(clf.score(bin_dense, bin_target), 1.0)
assert_array_almost_equal(1 - y * np.dot(K, clf.coef_.ravel()),
clf.errors_.ravel())
def test_fit_linear_binary_l2r_modified_huber():
clf = PrimalLinearSVC(C=1.0, random_state=0, penalty="l2",
loss="modified_huber")
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 1.0)
def test_fit_linear_binary_l2r_log():
clf = PrimalLinearSVC(C=1.0, random_state=0, penalty="l2", loss="log",
max_iter=5)
clf.fit(bin_dense, bin_target)
acc = clf.score(bin_dense, bin_target)
assert_almost_equal(acc, 1.0)
