def test_raise_exceptions(self, ):
# test that we raise errors in mishandled situations.
n_features = 2
n = 10
random_state = 123
X, y, _ = self._get_policy_data(n, n_features, random_state)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=20, max_samples=20).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=20, max_samples=1.2).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4,
criterion='peculiar').fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4, max_depth=-1).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4,
min_samples_split=-1).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4,
min_samples_leaf=-1).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4,
min_weight_fraction_leaf=-1.0).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4, max_features=10).fit(X, y)
with np.testing.assert_raises(ValueError):
forest = PolicyForest(n_estimators=4,
min_balancedness_tol=.55).fit(X, y)
return
def test_pickling(self, ):
n_features = 2
n = 10
random_state = 123
X, y, _ = self._get_policy_data(n, n_features, random_state)
forest = PolicyForest(n_estimators=4,
warm_start=True,
random_state=123).fit(X, y)
forest.n_estimators = 8
forest.fit(X, y)
pred1 = forest.predict(X)
joblib.dump(forest, 'forest.jbl')
loaded_forest = joblib.load('forest.jbl')
np.testing.assert_equal(loaded_forest.n_estimators,
forest.n_estimators)
np.testing.assert_allclose(loaded_forest.predict(X), pred1)
def _train_policy_forest(self, X, y, config, sample_weight=None):
return PolicyForest(**config).fit(X, y, sample_weight=sample_weight)
def test_warm_start(self, ):
n_features = 2
n = 10
random_state = 123
X, y, _ = self._get_policy_data(n, n_features, random_state)
forest = PolicyForest(n_estimators=4,
warm_start=True,
random_state=123).fit(X, y)
with pytest.warns(UserWarning):
forest.fit(X, y)
forest.n_estimators = 3
with np.testing.assert_raises(ValueError):
forest.fit(X, y)
forest.n_estimators = 8
forest.fit(X, y)
pred1 = forest.predict(X)
inds1 = forest.get_subsample_inds()
tree_states1 = [t.random_state for t in forest]
forest = PolicyForest(n_estimators=8,
warm_start=True,
random_state=123).fit(X, y)
pred2 = forest.predict(X)
inds2 = forest.get_subsample_inds()
tree_states2 = [t.random_state for t in forest]
np.testing.assert_allclose(pred1, pred2)
np.testing.assert_allclose(inds1, inds2)
np.testing.assert_allclose(tree_states1, tree_states2)
return
def test_non_standard_input(self, ):
# test that the estimator accepts lists, tuples and pandas data frames
n_features = 2
n = 2000
random_state = 123
X, y, _ = self._get_policy_data(n, n_features, random_state)
forest = PolicyForest(n_estimators=20, n_jobs=1,
random_state=123).fit(X, y)
pred = forest.predict(X)
pred_val = forest.predict_value(X)
pred_prob = forest.predict_proba(X)
assert pred_prob.shape == (X.shape[0], 2)
feat_imp = forest.feature_importances()
forest = PolicyForest(n_estimators=20, n_jobs=1,
random_state=123).fit(X.astype(np.float32),
np.asfortranarray(y))
np.testing.assert_allclose(pred, forest.predict(tuple(X)))
np.testing.assert_allclose(pred_val, forest.predict_value(tuple(X)))
forest = PolicyForest(n_estimators=20, n_jobs=1,
random_state=123).fit(tuple(X), tuple(y))
np.testing.assert_allclose(pred, forest.predict(tuple(X)))
np.testing.assert_allclose(pred_val, forest.predict_value(tuple(X)))
np.testing.assert_allclose(pred_prob, forest.predict_proba(tuple(X)))
forest = PolicyForest(n_estimators=20, n_jobs=1,
random_state=123).fit(list(X), list(y))
np.testing.assert_allclose(pred, forest.predict(list(X)))
np.testing.assert_allclose(pred_val, forest.predict_value(list(X)))
np.testing.assert_allclose(pred_prob, forest.predict_proba(list(X)))
forest = PolicyForest(n_estimators=20, n_jobs=1,
random_state=123).fit(pd.DataFrame(X),
pd.DataFrame(y))
np.testing.assert_allclose(pred, forest.predict(pd.DataFrame(X)))
np.testing.assert_allclose(pred_val,
forest.predict_value(pd.DataFrame(X)))
np.testing.assert_allclose(pred_prob,
forest.predict_proba(pd.DataFrame(X)))
groups = np.repeat(np.arange(X.shape[0]), 2)
Xraw = X.copy()
X = np.repeat(X, 2, axis=0)
T = np.zeros(y.shape)
T[:, 1] = 1
T = T.flatten()
y = y.flatten()
forest = DRPolicyForest(
model_regression=DummyRegressor(strategy='constant', constant=0),
model_propensity=DummyClassifier(strategy='uniform'),
featurizer=PolynomialFeatures(degree=1, include_bias=False),
cv=GroupKFold(n_splits=2),
n_estimators=20,
n_jobs=1,
random_state=123).fit(y, T, X=X, groups=groups)
mask = np.abs(Xraw[:, 0]) > .1
np.testing.assert_allclose(pred[mask], forest.predict(Xraw[mask]))
np.testing.assert_allclose(pred_val[mask, 1] - pred_val[mask, 0],
forest.predict_value(Xraw[mask]).flatten(),
atol=.08)
np.testing.assert_allclose(feat_imp,
forest.feature_importances(),
atol=1e-4)
np.testing.assert_allclose(feat_imp,
forest.feature_importances_,
atol=1e-4)
pred = forest.predict(X)
pred_val = forest.predict_value(X)
pred_prob = forest.predict_proba(X)
np.testing.assert_allclose(pred, forest.predict(tuple(X)))
np.testing.assert_allclose(pred_val, forest.predict_value(tuple(X)))
np.testing.assert_allclose(pred, forest.predict(pd.DataFrame(X)))
np.testing.assert_allclose(pred_val,
forest.predict_value(pd.DataFrame(X)))
np.testing.assert_allclose(pred_prob,
forest.predict_proba(pd.DataFrame(X)))
return