def test_anomaly_tuning():
"""Check anomaly_tuning gives same results than est_tuning."""
parameters = {'k': np.arange(1, 10), 'novelty': [True]}
alphas = np.array([0.2, 0.5, 0.9])
models, offsets = anomaly_tuning(X,
AverageKLPE,
alphas=alphas,
parameters=parameters,
cv=cv,
n_sim=100)
score_estimators = np.empty((n_estimator, len(X)))
for i in range(n_estimator):
score_estimators[i, :] = models[i].score_samples(X)
score_estimators_seq = np.empty((n_estimator, len(X)))
offsets_seq = np.empty((n_estimator, len(alphas)))
param_grid = ParameterGrid(parameters)
i = 0
for train, test in cv.split(X):
X_train = X[train]
X_test = X[test]
model, offset = est_tuning(X_train, X_test, AverageKLPE, param_grid,
alphas, U, vol_tot_cube)
score_estimators_seq[i, :] = model.score_samples(X)
offsets_seq[i, :] = offset
i += 1
assert_array_almost_equal(score_estimators, score_estimators_seq)
assert_array_almost_equal(offsets, offsets_seq)
def test_est_tuning():
"""Check that est_tuning returns the estimator with minimum auc."""
for algo in algorithms:
name_algo = algo.name
parameters = algo_param[name_algo]
param_grid = ParameterGrid(parameters)
alphas = rng.randint(1, 100, size=5) / 100
alphas = np.sort(alphas)
clf_est, offsets_est = est_tuning(X_train, X_test, algo, param_grid,
alphas, U, vol_tot_cube)
# check that clf_est gives the minimum auc
score_function = clf_est.score_samples
vol_est, _ = _compute_volumes(score_function, alphas, X_test, U,
vol_tot_cube)
auc_est = auc(alphas, vol_est)
auc_algo = np.zeros(len(param_grid))
for p, param in enumerate(param_grid):
clf = algo(**param)
clf = clf.fit(X_train)
score_function_p = clf.score_samples
vol_p, _ = _compute_volumes(score_function_p, alphas, X_test, U,
vol_tot_cube)
auc_algo[p] = auc(alphas, vol_p)
assert_equal(np.min(auc_algo), auc_est)
clf_test = clf_est.score_samples(X_test)
proba_offsets_est = (clf_test >= offsets_est[:, np.newaxis])
# this test requires to have a large number of samples because
# np.percentile is an empirical quantile which uses interpolation.
# this is also why we ask the values to be equal only up to the
# second decimal.
assert_array_almost_equal(np.mean(proba_offsets_est, axis=1),
alphas,
decimal=2)