def test_fit_exception(self):
with self.assertRaises(Exception) as cm:
pmml = path.join(BASE_DIR, '../models/categorical-rf.pmml')
clf = PMMLForestClassifier(pmml)
clf.fit(np.array([[]]), np.array([]))
assert str(cm.exception) == 'Not supported.'
def setUp(self):
df = pd.read_csv(path.join(BASE_DIR, '../models/categorical-test.csv'))
cats = np.unique(df['age'])
df['age'] = pd.Categorical(df['age'], categories=cats)
Xte = df.iloc[:, 1:]
yte = df.iloc[:, 0]
self.test = (Xte, yte)
pmml = path.join(BASE_DIR, '../models/categorical-rf.pmml')
self.clf = PMMLForestClassifier(pmml)
def setUp(self):
from sklearn.datasets import load_iris
iris = load_iris()
X = pd.DataFrame(iris.data)
X.columns = np.array(['sepal_length', 'sepal_width', 'petal_length',
'petal_width'])
y = pd.Series(np.array(np.array(['setosa', 'versicolor', 'virginica']))[iris.target])
y.name = "Class"
self.test = X, y
self.clf = PMMLForestClassifier(path.join(BASE_DIR, '../models/randomForest.pmml'))
self.ref = RandomForestClassifier(random_state=1).fit(X, y)
def test_non_true_segment(self):
with self.assertRaises(Exception), catch_warnings(record=True) as w:
PMMLForestClassifier(pmml=StringIO("""
<PMML xmlns="http://www.dmg.org/PMML-4_3" version="4.3">
<DataDictionary>
<DataField name="Class" optype="categorical" dataType="string">
<Value value="setosa"/>
<Value value="versicolor"/>
<Value value="virginica"/>
</DataField>
</DataDictionary>
<MiningModel>
<MiningSchema>
<MiningField name="Class" usageType="target"/>
</MiningSchema>
<Segmentation multipleModelMethod="majorityVote">
<Segment>
<False/>
</Segment>
<Segment>
<True/>
</Segment>
</Segmentation>
</MiningModel>
</PMML>
"""))
assert len(w) == 1
def test_non_binary_tree(self):
with self.assertRaises(Exception) as cm:
PMMLForestClassifier(pmml=StringIO("""
<PMML xmlns="http://www.dmg.org/PMML-4_3" version="4.3">
<DataDictionary>
<DataField name="Class" optype="categorical" dataType="string">
<Value value="setosa"/>
<Value value="versicolor"/>
<Value value="virginica"/>
</DataField>
</DataDictionary>
<MiningModel>
<MiningSchema>
<MiningField name="Class" usageType="target"/>
</MiningSchema>
<Segmentation multipleModelMethod="majorityVote">
<Segment>
<True/>
<TreeModel splitCharacteristic="multiSplit" />
</Segment>
</Segmentation>
</MiningModel>
</PMML>
"""))
assert str(cm.exception) == 'Sklearn only supports binary tree models.'
def test_sklearn2pmml(self):
# Export to PMML
pipeline = PMMLPipeline([("classifier", self.ref)])
pipeline.fit(self.test[0], self.test[1])
sklearn2pmml(pipeline, "rf-sklearn2pmml.pmml", with_repr=True)
try:
# Import PMML
model = PMMLForestClassifier(pmml='rf-sklearn2pmml.pmml')
# Verify classification
Xte, _ = self.test
assert np.array_equal(self.ref.predict_proba(Xte),
model.predict_proba(Xte))
finally:
remove("rf-sklearn2pmml.pmml")
def test_invalid_model(self):
with self.assertRaises(Exception) as cm:
PMMLForestClassifier(pmml=StringIO("""
<PMML xmlns="http://www.dmg.org/PMML-4_3" version="4.3">
<DataDictionary>
<DataField name="Class" optype="categorical" dataType="string">
<Value value="setosa"/>
<Value value="versicolor"/>
<Value value="virginica"/>
</DataField>
</DataDictionary>
<MiningSchema>
<MiningField name="Class" usageType="target"/>
</MiningSchema>
</PMML>
"""))
assert str(cm.exception) == 'PMML model does not contain MiningModel.'
def test_non_voting_ensemble(self):
with self.assertRaises(Exception) as cm:
PMMLForestClassifier(pmml=StringIO("""
<PMML xmlns="http://www.dmg.org/PMML-4_3" version="4.3">
<DataDictionary>
<DataField name="Class" optype="categorical" dataType="string">
<Value value="setosa"/>
<Value value="versicolor"/>
<Value value="virginica"/>
</DataField>
</DataDictionary>
<MiningModel>
<MiningSchema>
<MiningField name="Class" usageType="target"/>
</MiningSchema>
<Segmentation multipleModelMethod="mean" />
</MiningModel>
</PMML>
"""))
assert str(cm.exception) == 'PMML model ensemble should use majority vote or average.'
class TestCategoricalPimaForestIntegration(TestCase):
def setUp(self):
df = pd.read_csv(path.join(BASE_DIR, '../models/categorical-test.csv'))
cats = np.unique(df['age'])
df['age'] = pd.Categorical(df['age'], categories=cats)
Xte = df.iloc[:, 1:]
yte = df.iloc[:, 0]
self.test = (Xte, yte)
pmml = path.join(BASE_DIR, '../models/categorical-rf.pmml')
self.clf = PMMLForestClassifier(pmml)
def test_predict_proba(self):
Xte, _ = self.test
ref = np.array([
[0.2200000000000000, 0.7800000000000000],
[0.5800000000000000, 0.4200000000000000],
[0.2200000000000000, 0.7800000000000000],
[0.1800000000000000, 0.8200000000000000],
[0.0800000000000000, 0.9200000000000000],
[0.3000000000000000, 0.7000000000000000],
[0.2600000000000000, 0.7400000000000000],
[0.3600000000000000, 0.6400000000000000],
[0.4000000000000000, 0.6000000000000000],
[0.1600000000000000, 0.8400000000000000],
[0.2200000000000000, 0.7800000000000000],
[0.3000000000000000, 0.7000000000000000],
[0.1400000000000000, 0.8600000000000000],
[0.6400000000000000, 0.3600000000000000],
[0.1800000000000000, 0.8200000000000000],
[0.1800000000000000, 0.8200000000000000],
[0.7600000000000000, 0.2400000000000000],
[0.2400000000000000, 0.7600000000000000],
[0.3400000000000000, 0.6600000000000000],
[0.2800000000000000, 0.7200000000000000],
[0.0800000000000000, 0.9200000000000000],
[0.2000000000000000, 0.8000000000000000],
[0.6800000000000000, 0.3200000000000000],
[0.1200000000000000, 0.8800000000000000],
[0.2200000000000000, 0.7800000000000000],
[0.3600000000000000, 0.6400000000000000],
[0.2000000000000000, 0.8000000000000000],
[0.8600000000000000, 0.1400000000000000],
[0.9399999999999999, 0.0600000000000000],
[0.7200000000000000, 0.2800000000000000],
[0.5600000000000001, 0.4400000000000000],
[0.9800000000000000, 0.0200000000000000],
[0.4400000000000000, 0.5600000000000001],
[0.8800000000000000, 0.1200000000000000],
[0.6600000000000000, 0.3400000000000000],
[0.5000000000000000, 0.5000000000000000],
[0.7400000000000000, 0.2600000000000000],
[0.2600000000000000, 0.7400000000000000],
[0.1600000000000000, 0.8400000000000000],
[0.6800000000000000, 0.3200000000000000],
[0.7600000000000000, 0.2400000000000000],
[0.7400000000000000, 0.2600000000000000],
[0.5600000000000001, 0.4400000000000000],
[0.5400000000000000, 0.4600000000000000],
[0.5200000000000000, 0.4800000000000000],
[0.1400000000000000, 0.8600000000000000],
[0.7600000000000000, 0.2400000000000000],
[0.8200000000000000, 0.1800000000000000],
[0.4400000000000000, 0.5600000000000001],
[0.9200000000000000, 0.0800000000000000],
[0.5600000000000001, 0.4400000000000000],
[0.2800000000000000, 0.7200000000000000]
])
assert np.allclose(ref, self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
ref = 0.7884615384615384
assert ref == self.clf.score(Xte, yte)
class TestIrisForestIntegration(TestCase):
def setUp(self):
from sklearn.datasets import load_iris
iris = load_iris()
X = pd.DataFrame(iris.data)
X.columns = np.array(['sepal_length', 'sepal_width', 'petal_length',
'petal_width'])
y = pd.Series(np.array(np.array(['setosa', 'versicolor', 'virginica']))[iris.target])
y.name = "Class"
self.test = X, y
self.clf = PMMLForestClassifier(path.join(BASE_DIR, '../models/randomForest.pmml'))
self.ref = RandomForestClassifier(random_state=1).fit(X, y)
def test_predict_proba(self):
Xte, _ = self.test
ref = np.array([
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.985, 0.015, 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.99, 0.01, 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.98, 0.02, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.98, 0.015, 0.005],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.595, 0.405],
[0., 0.995, 0.005],
[0., 0.99, 0.01],
[0., 0.995, 0.005],
[0., 0.97, 0.03],
[0.005, 0.805, 0.19],
[0., 1., 0.],
[0.005, 0.985, 0.01],
[0., 0.95, 0.05],
[0., 0.995, 0.005],
[0., 0.985, 0.015],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.99, 0.01],
[0., 0.995, 0.005],
[0., 0.645, 0.355],
[0., 1., 0.],
[0., 0.76, 0.24],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.85, 0.15],
[0., 0.025, 0.975],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.085, 0.915],
[0.025, 0.96, 0.015],
[0.01, 0.945, 0.045],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 0.98, 0.02],
[0., 0.995, 0.005],
[0., 0.99, 0.01],
[0., 0.955, 0.045],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.99, 0.01],
[0., 1., 0.],
[0., 0., 1.],
[0., 0.01, 0.99],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.245, 0.755],
[0., 0., 1.],
[0., 0.065, 0.935],
[0., 0., 1.],
[0., 0.01, 0.99],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.055, 0.945],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.015, 0.985],
[0., 0.19, 0.81],
[0., 0., 1.],
[0., 0.11, 0.89],
[0., 0., 1.],
[0., 0.07, 0.93],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.12, 0.88],
[0., 0.035, 0.965],
[0., 0., 1.],
[0., 0.05, 0.95],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0.115, 0.885],
[0., 0.09, 0.91],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.18, 0.82],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.01, 0.99],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.09, 0.91],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.025, 0.975]
])
assert np.array_equal(ref, self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
ref = 0.9866666666666667
assert ref == self.clf.score(Xte, yte)
def test_sklearn2pmml(self):
# Export to PMML
pipeline = PMMLPipeline([
("classifier", self.ref)
])
pipeline.fit(self.test[0], self.test[1])
sklearn2pmml(pipeline, "forest_sklearn2pmml.pmml", with_repr = True)
try:
# Import PMML
model = PMMLForestClassifier(pmml='forest_sklearn2pmml.pmml')
# Verify classification
Xte, _ = self.test
assert np.array_equal(
self.ref.predict_proba(Xte),
model.predict_proba(Xte)
)
finally:
remove("forest_sklearn2pmml.pmml")
class TestIrisForestIntegration(TestCase):
def setUp(self):
from sklearn.datasets import load_iris
iris = load_iris()
X = pd.DataFrame(iris.data)
X.columns = np.array(['sepal_length', 'sepal_width', 'petal_length',
'petal_width'])
y = pd.Series(np.array(np.array(['Iris-setosa', 'Iris-versicolor',
'Iris-virginica']))[iris.target])
y.name = "Class"
self.test = X, y
self.clf = PMMLForestClassifier(path.join(BASE_DIR, '../models/iris.pmml'))
def test_predict_proba(self):
Xte, _ = self.test
ref = np.array([
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.985, 0.015, 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.99, 0.01, 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.995, 0.005, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.98, 0.02, 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0.98, 0.015, 0.005],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.595, 0.405],
[0., 0.995, 0.005],
[0., 0.99, 0.01],
[0., 0.995, 0.005],
[0., 0.97, 0.03],
[0.005, 0.805, 0.19],
[0., 1., 0.],
[0.005, 0.985, 0.01],
[0., 0.95, 0.05],
[0., 0.995, 0.005],
[0., 0.985, 0.015],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.99, 0.01],
[0., 0.995, 0.005],
[0., 0.645, 0.355],
[0., 1., 0.],
[0., 0.76, 0.24],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.85, 0.15],
[0., 0.025, 0.975],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.085, 0.915],
[0.025, 0.96, 0.015],
[0.01, 0.945, 0.045],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 1., 0.],
[0., 0.995, 0.005],
[0., 0.98, 0.02],
[0., 0.995, 0.005],
[0., 0.99, 0.01],
[0., 0.955, 0.045],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 1., 0.],
[0., 0.99, 0.01],
[0., 1., 0.],
[0., 0., 1.],
[0., 0.01, 0.99],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.245, 0.755],
[0., 0., 1.],
[0., 0.065, 0.935],
[0., 0., 1.],
[0., 0.01, 0.99],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.055, 0.945],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.015, 0.985],
[0., 0.19, 0.81],
[0., 0., 1.],
[0., 0.11, 0.89],
[0., 0., 1.],
[0., 0.07, 0.93],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.12, 0.88],
[0., 0.035, 0.965],
[0., 0., 1.],
[0., 0.05, 0.95],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0.115, 0.885],
[0., 0.09, 0.91],
[0., 0., 1.],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.18, 0.82],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.005, 0.995],
[0., 0.01, 0.99],
[0., 0.005, 0.995],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.09, 0.91],
[0., 0., 1.],
[0., 0., 1.],
[0., 0.025, 0.975]
])
assert np.array_equal(ref, self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
ref = 0.9866666666666667
assert ref == self.clf.score(Xte, yte)
def test_more_tags(self):
clf = PMMLForestClassifier(
path.join(BASE_DIR, '../models/rf-cat-pima.pmml'))
assert clf._more_tags() == RandomForestClassifier()._more_tags()