class TestDigitsTreeIntegration(TestCase):
def setUp(self):
data = load_digits()
self.columns = [2, 3, 4, 5, 6, 7, 9, 10, 13, 14, 17, 18, 19, 20, 21, 25, 26,
27, 28, 29, 30, 33, 34, 35, 36, 37, 38, 41, 42, 43, 45, 46,
50, 51, 52, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63]
X = pd.DataFrame(data.data)
y = pd.Series(np.array(data.target_names)[data.target])
y.name = "Class"
X, Xte, y, yte = train_test_split(X, y, test_size=0.33, random_state=123)
self.test = (Xte, yte)
self.clf = PMMLTreeClassifier(path.join(BASE_DIR, '../models/digits.pmml'))
self.ref = DecisionTreeClassifier(random_state=1).fit(X, y)
def test_predict(self):
Xte, _ = self.test
assert np.array_equal(
self.ref.predict(Xte),
self.clf.predict(Xte[self.columns]).astype(np.int64)
)
def test_predict_proba(self):
Xte, _ = self.test
assert np.array_equal(
self.ref.predict_proba(Xte),
self.clf.predict_proba(Xte[self.columns])
)
def test_score(self):
Xte, yte = self.test
assert self.ref.score(Xte, yte) == self.clf.score(Xte[self.columns], yte)
class TestIrisTreeIntegration(TestCase):
def setUp(self):
pair = [0, 1]
data = load_iris()
X = pd.DataFrame(data.data[:, pair])
X.columns = np.array(data.feature_names)[pair]
y = pd.Series(np.array(data.target_names)[data.target])
y.name = "Class"
X, Xte, y, yte = train_test_split(X, y, test_size=0.33, random_state=123)
self.test = (Xte, yte)
pmml = path.join(BASE_DIR, '../models/sklearn2pmml.pmml')
self.clf = PMMLTreeClassifier(pmml=pmml)
self.ref = DecisionTreeClassifier(random_state=1).fit(X, y)
def test_predict(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict(Xte), self.clf.predict(Xte))
def test_predict_proba(self):
Xte, _ = self.test
assert np.array_equal(
self.ref.predict_proba(Xte),
self.clf.predict_proba(Xte)
)
def test_score(self):
Xte, yte = self.test
assert self.ref.score(Xte, yte) == self.clf.score(Xte, yte)
class TestDigitsTreeIntegration(TestCase):
def setUp(self):
data = load_digits()
X = pd.DataFrame(data.data)
y = pd.Series(np.array(data.target_names)[data.target])
y.name = "Class"
X, Xte, y, yte = train_test_split(X,
y,
test_size=0.33,
random_state=123)
self.test = (Xte, yte)
self.clf = PMMLTreeClassifier(
path.join(BASE_DIR, '../models/tree-digits.pmml'))
self.ref = DecisionTreeClassifier(random_state=1).fit(X, y)
def test_predict(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict(Xte), self.clf.predict(Xte))
def test_predict_proba(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict_proba(Xte),
self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
assert self.ref.score(Xte, yte) == self.clf.score(Xte, yte)
class TestIrisTreeIntegration(TestCase):
def setUp(self):
pair = [0, 1]
data = load_iris()
X = pd.DataFrame(data.data[:, pair])
X.columns = np.array(data.feature_names)[pair]
y = pd.Series(np.array(data.target_names)[data.target])
y.name = "Class"
X, Xte, y, yte = train_test_split(X, y, test_size=0.33, random_state=123)
self.test = (Xte, yte)
self.train = (X, y)
pmml = path.join(BASE_DIR, '../models/decisionTree.pmml')
self.clf = PMMLTreeClassifier(pmml=pmml)
self.ref = DecisionTreeClassifier(random_state=1).fit(X, y)
def test_predict(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict(Xte), self.clf.predict(Xte))
def test_predict_proba(self):
Xte, _ = self.test
assert np.array_equal(
self.ref.predict_proba(Xte),
self.clf.predict_proba(Xte)
)
def test_score(self):
Xte, yte = self.test
assert self.ref.score(Xte, yte) == self.clf.score(Xte, yte)
def test_sklearn2pmml(self):
# Export to PMML
pipeline = PMMLPipeline([
("classifier", self.ref)
])
pipeline.fit(self.train[0], self.train[1])
sklearn2pmml(pipeline, "tree_sklearn2pmml.pmml", with_repr = True)
try:
# Import PMML
model = PMMLTreeClassifier(pmml='tree_sklearn2pmml.pmml')
# Verify classification
Xte, _ = self.test
assert np.array_equal(
self.ref.predict_proba(Xte),
model.predict_proba(Xte)
)
finally:
remove("tree_sklearn2pmml.pmml")
class TestIrisTreeIntegration(TestCase):
def setUp(self):
pair = [0, 1]
data = load_iris(as_frame=True)
X = data.data
y = data.target
y.name = "Class"
self.test = (X, y)
pmml = path.join(BASE_DIR, '../models/tree-iris.pmml')
self.clf = PMMLTreeClassifier(pmml=pmml)
self.ref = DecisionTreeClassifier(random_state=1).fit(X, y)
def test_predict(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict(Xte), self.clf.predict(Xte))
def test_predict_proba(self):
Xte, _ = self.test
assert np.array_equal(self.ref.predict_proba(Xte),
self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
assert self.ref.score(Xte, yte) == 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, "tree-sklearn2pmml.pmml", with_repr=True)
try:
# Import PMML
model = PMMLTreeClassifier(pmml='tree-sklearn2pmml.pmml')
# Verify classification
Xte, _ = self.test
assert np.array_equal(self.ref.predict_proba(Xte),
model.predict_proba(Xte))
finally:
remove("tree-sklearn2pmml.pmml")
class TestCategoricalPimaTreeIntegration(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)
df['age'] = df['age'].cat.codes
Xte = df.iloc[:, 1:]
yte = df.iloc[:, 0]
self.test = (Xte, yte)
pmml = path.join(BASE_DIR, '../models/categorical.pmml')
self.clf = PMMLTreeClassifier(pmml)
def test_predict_proba(self):
Xte, _ = self.test
ref = np.array([
[0.1568627450980392, 0.84313725490196079],
[0.7500000000000000, 0.25000000000000000],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.2000000000000000, 0.80000000000000004],
[0.2000000000000000, 0.80000000000000004],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.2000000000000000, 0.80000000000000004],
[0.9428571428571428, 0.05714285714285714],
[0.2000000000000000, 0.80000000000000004],
[0.2000000000000000, 0.80000000000000004],
[0.9428571428571428, 0.05714285714285714],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.7368421052631579, 0.26315789473684209],
[0.1568627450980392, 0.84313725490196079],
[0.2000000000000000, 0.80000000000000004],
[0.2000000000000000, 0.80000000000000004],
[0.2000000000000000, 0.80000000000000004],
[0.9428571428571428, 0.05714285714285714],
[0.9428571428571428, 0.05714285714285714],
[0.7368421052631579, 0.26315789473684209],
[0.9428571428571428, 0.05714285714285714],
[0.9428571428571428, 0.05714285714285714],
[0.7368421052631579, 0.26315789473684209],
[0.9428571428571428, 0.05714285714285714],
[0.7368421052631579, 0.26315789473684209],
[0.7500000000000000, 0.25000000000000000],
[0.7368421052631579, 0.26315789473684209],
[0.1568627450980392, 0.84313725490196079],
[0.2000000000000000, 0.80000000000000004],
[0.7368421052631579, 0.26315789473684209],
[0.9428571428571428, 0.05714285714285714],
[0.9428571428571428, 0.05714285714285714],
[0.1568627450980392, 0.84313725490196079],
[0.7368421052631579, 0.26315789473684209],
[0.1568627450980392, 0.84313725490196079],
[0.1568627450980392, 0.84313725490196079],
[0.7368421052631579, 0.26315789473684209],
[0.7368421052631579, 0.26315789473684209],
[0.1568627450980392, 0.84313725490196079],
[0.9428571428571428, 0.05714285714285714],
[0.7368421052631579, 0.26315789473684209],
[0.2000000000000000, 0.80000000000000004]
])
assert np.array_equal(ref, self.clf.predict_proba(Xte))
def test_score(self):
Xte, yte = self.test
ref = 0.7692307692307693
assert ref == self.clf.score(Xte, yte)