def sdml_fit(samples, similarity_set, prior='covariance', balance_param=0.15):
"""Prior can be 'covariance', 'identity' or 'random'.
balance_param was used 0.5 in the first version of the paper, but it does not work here with such a large value. """
n_samples = len(similarity_set)
sdml = SDML(prior=prior,
preprocessor=samples,
verbose=True,
balance_param=balance_param)
pairs, Y = [], []
for ind1 in range(n_samples):
for ind2 in range(n_samples):
pairs.append([ind1, ind2])
if similarity_set[ind1, ind2]:
Y.append(1)
else:
Y.append(-1)
start = time()
sdml.fit(pairs, Y)
print("Fitting took {:.2f} seconds.".format(time() - start))
return sdml
def test_sdml_converges_if_psd(self):
"""Tests that sdml converges on a simple problem where we know the
pseudo-covariance matrix is PSD"""
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]])
y = [1, -1]
sdml = SDML(use_cov=True, sparsity_param=0.01, balance_param=0.5)
sdml.fit(pairs, y)
assert np.isfinite(sdml.get_mahalanobis_matrix()).all()
def test_sdml_converges_if_psd(self): """Tests that sdml converges on a simple problem where we know the pseudo-covariance matrix is PSD""" pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]]) y = [1, -1] sdml = SDML(use_cov=True, sparsity_param=0.01, balance_param=0.5) sdml.fit(pairs, y) assert np.isfinite(sdml.get_mahalanobis_matrix()).all()
def test_verbose_has_not_installed_skggm_sdml(capsys):
# Test that if users have installed skggm, a message is printed telling them
# skggm's solver is used (when they use SDML)
# TODO: remove if we don't need skggm anymore
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]])
y_pairs = [1, -1]
sdml = SDML(verbose=True)
sdml.fit(pairs, y_pairs)
out, _ = capsys.readouterr()
assert "SDML will use scikit-learn's graphical lasso solver." in out
def test_verbose_has_not_installed_skggm_sdml(capsys): # Test that if users have installed skggm, a message is printed telling them # skggm's solver is used (when they use SDML) # TODO: remove if we don't need skggm anymore pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]]) y_pairs = [1, -1] sdml = SDML(verbose=True) sdml.fit(pairs, y_pairs) out, _ = capsys.readouterr() assert "SDML will use scikit-learn's graphical lasso solver." in out
def test_raises_no_warning_installed_skggm(self):
# otherwise we should be able to instantiate and fit SDML and it
# should raise no warning
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]])
y_pairs = [1, -1]
X, y = make_classification(random_state=42)
with pytest.warns(None) as record:
sdml = SDML()
sdml.fit(pairs, y_pairs)
assert len(record) == 0
with pytest.warns(None) as record:
sdml = SDML_Supervised(use_cov=False, balance_param=1e-5)
sdml.fit(X, y)
assert len(record) == 0
def test_raises_no_warning_installed_skggm(self):
# otherwise we should be able to instantiate and fit SDML and it
# should raise no warning
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]])
y_pairs = [1, -1]
X, y = make_classification(random_state=42)
with pytest.warns(None) as record:
sdml = SDML()
sdml.fit(pairs, y_pairs)
assert len(record) == 0
with pytest.warns(None) as record:
sdml = SDML_Supervised(use_cov=False, balance_param=1e-5)
sdml.fit(X, y)
assert len(record) == 0
def test_sdml_raises_warning_msg_installed_skggm(self):
"""Tests that the right warning message is raised if someone tries to
use SDML but has not installed skggm, and that the algorithm fails to
converge"""
# TODO: remove if we don't need skggm anymore
# case on which we know that skggm's graphical lasso fails
# because it will return non finite values
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y_pairs = [1, -1]
sdml = SDML(use_cov=False, balance_param=100, verbose=True)
msg = ("There was a problem in SDML when using skggm's graphical "
"lasso solver.")
with pytest.raises(RuntimeError) as raised_error:
sdml.fit(pairs, y_pairs)
assert msg == str(raised_error.value)
def test_raises_no_warning_installed_skggm(self):
# otherwise we should be able to instantiate and fit SDML and it
# should raise no error and no ConvergenceWarning
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., -55.], [0., -60]]])
y_pairs = [1, -1]
X, y = make_classification(random_state=42)
with pytest.warns(None) as records:
sdml = SDML(prior='covariance')
sdml.fit(pairs, y_pairs)
for record in records:
assert record.category is not ConvergenceWarning
with pytest.warns(None) as records:
sdml_supervised = SDML_Supervised(prior='identity', balance_param=1e-5)
sdml_supervised.fit(X, y)
for record in records:
assert record.category is not ConvergenceWarning
def test_sdml_raises_warning_msg_installed_skggm(self):
"""Tests that the right warning message is raised if someone tries to
use SDML but has not installed skggm, and that the algorithm fails to
converge"""
# TODO: remove if we don't need skggm anymore
# case on which we know that skggm's graphical lasso fails
# because it will return non finite values
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y_pairs = [1, -1]
sdml = SDML(use_cov=False, balance_param=100, verbose=True)
msg = ("There was a problem in SDML when using skggm's graphical "
"lasso solver.")
with pytest.raises(RuntimeError) as raised_error:
sdml.fit(pairs, y_pairs)
assert msg == str(raised_error.value)
def test_sdml_raises_warning_non_psd(self):
"""Tests that SDML raises a warning on a toy example where we know the
pseudo-covariance matrix is not PSD"""
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y = [1, -1]
sdml = SDML(use_cov=True, sparsity_param=0.01, balance_param=0.5)
msg = ("Warning, the input matrix of graphical lasso is not "
"positive semi-definite (PSD). The algorithm may diverge, "
"and lead to degenerate solutions. "
"To prevent that, try to decrease the balance parameter "
"`balance_param` and/or to set use_cov=False.")
with pytest.warns(ConvergenceWarning) as raised_warning:
try:
sdml.fit(pairs, y)
except Exception:
pass
# we assert that this warning is in one of the warning raised by the
# estimator
assert msg in list(map(lambda w: str(w.message), raised_warning))
def test_sdml_raises_warning_msg_not_installed_skggm(self):
"""Tests that the right warning message is raised if someone tries to
use SDML but has not installed skggm, and that the algorithm fails to
converge"""
# TODO: remove if we don't need skggm anymore
# case on which we know that scikit-learn's graphical lasso fails
# because it will return a non SPD matrix
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y_pairs = [1, -1]
sdml = SDML(prior='identity', balance_param=100, verbose=True)
msg = ("There was a problem in SDML when using scikit-learn's graphical "
"lasso solver. skggm's graphical lasso can sometimes converge on "
"non SPD cases where scikit-learn's graphical lasso fails to "
"converge. Try to install skggm and rerun the algorithm (see "
"the README.md for the right version of skggm).")
with pytest.raises(RuntimeError) as raised_error:
sdml.fit(pairs, y_pairs)
assert msg == str(raised_error.value)
def test_sdml_raises_warning_non_psd(self):
"""Tests that SDML raises a warning on a toy example where we know the
pseudo-covariance matrix is not PSD"""
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y = [1, -1]
sdml = SDML(use_cov=True, sparsity_param=0.01, balance_param=0.5)
msg = ("Warning, the input matrix of graphical lasso is not "
"positive semi-definite (PSD). The algorithm may diverge, "
"and lead to degenerate solutions. "
"To prevent that, try to decrease the balance parameter "
"`balance_param` and/or to set use_cov=False.")
with pytest.warns(ConvergenceWarning) as raised_warning:
try:
sdml.fit(pairs, y)
except Exception:
pass
# we assert that this warning is in one of the warning raised by the
# estimator
assert msg in list(map(lambda w: str(w.message), raised_warning))
def test_sdml_raises_warning_msg_not_installed_skggm(self):
"""Tests that the right warning message is raised if someone tries to
use SDML but has not installed skggm, and that the algorithm fails to
converge"""
# TODO: remove if we don't need skggm anymore
# case on which we know that scikit-learn's graphical lasso fails
# because it will return a non SPD matrix
pairs = np.array([[[-10., 0.], [10., 0.]], [[0., 50.], [0., -60]]])
y_pairs = [1, -1]
sdml = SDML(use_cov=False, balance_param=100, verbose=True)
msg = ("There was a problem in SDML when using scikit-learn's graphical "
"lasso solver. skggm's graphical lasso can sometimes converge on "
"non SPD cases where scikit-learn's graphical lasso fails to "
"converge. Try to install skggm and rerun the algorithm (see "
"the README.md for the right version of skggm).")
with pytest.raises(RuntimeError) as raised_error:
sdml.fit(pairs, y_pairs)
assert msg == str(raised_error.value)
def test_tiwafer():
num_constraints = 1500
print "Loading Data...."
tiwafer_data = load_data_sdml()
sim_pairs = tiwafer_data.sim_pairs
diff_pairs = tiwafer_data.diff_pairs
sorted_ids = tiwafer_data.sortedIds
ti_data = np.array(tiwafer_data.data)
labels = np.array(tiwafer_data.target)
print "Done Loading Data.\nLearning Distance Metric...."
num_points = len(sorted_ids)
W = prepare_constraints_old(labels, num_points, num_constraints)
sdml = SDML()
# W = prepare_constraints(sorted_ids, sim_pairs, diff_pairs)
sdml.fit(ti_data, W)
W_metric = sdml.metric()
cPickle.dump(W_metric, open('W_metric_sdml.p', 'wb'))
W_trans = sdml.transformer()
with open('W_trans_sdml.p', 'wb') as handle:
cPickle.dump(W_trans, handle)
