Merge 8f3e657 into d73048e

README.md

@@ -82,13 +82,13 @@ The output is consistent with the output of the `predict_proba` method of `Decis
 Here's a simple example of how a linear model trained in Python environment can be represented in Java code:
 ```python
 from sklearn.datasets import load_boston
-from sklearn import linear_model
+from sklearn.linear_model import LinearRegression
 import m2cgen as m2c
 
 boston = load_boston()
 X, y = boston.data, boston.target
 
-estimator = linear_model.LinearRegression()
+estimator = LinearRegression()
 estimator.fit(X, y)
 
 code = m2c.export_to_java(estimator)

tests/assemblers/test_ensemble.py

@@ -1,11 +1,11 @@
-from sklearn import ensemble
+from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier
 
 from m2cgen import assemblers, ast
 from tests import utils
 
 
 def test_single_condition():
-    estimator = ensemble.RandomForestRegressor(n_estimators=2, random_state=1)
+    estimator = RandomForestRegressor(n_estimators=2, random_state=1)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -30,7 +30,7 @@ def test_single_condition():
 
 
 def test_two_conditions():
-    estimator = ensemble.RandomForestRegressor(n_estimators=2, random_state=13)
+    estimator = RandomForestRegressor(n_estimators=2, random_state=13)
 
     estimator.fit([[1], [2], [3]], [1, 2, 3])
 
@@ -61,8 +61,7 @@ def test_two_conditions():
 
 
 def test_multi_class():
-    estimator = ensemble.RandomForestClassifier(
-        n_estimators=2, random_state=13)
+    estimator = RandomForestClassifier(n_estimators=2, random_state=13)
 
     estimator.fit([[1], [2], [3]], [1, -1, 1])
 

tests/assemblers/test_linear.py

@@ -4,7 +4,8 @@
 from statsmodels.regression.process_regression import ProcessMLE
 from lightning.regression import AdaGradRegressor
 from lightning.classification import AdaGradClassifier
-from sklearn import linear_model
+from sklearn.linear_model import \
+    LinearRegression, LogisticRegression, RANSACRegressor
 from sklearn.dummy import DummyRegressor
 from sklearn.tree import DecisionTreeRegressor
 
@@ -13,7 +14,7 @@
 
 
 def test_single_feature():
-    estimator = linear_model.LinearRegression()
+    estimator = LinearRegression()
     estimator.coef_ = np.array([1])
     estimator.intercept_ = np.array([3])
 
@@ -32,7 +33,7 @@ def test_single_feature():
 
 
 def test_two_features():
-    estimator = linear_model.LinearRegression()
+    estimator = LinearRegression()
     estimator.coef_ = np.array([1, 2])
     estimator.intercept_ = np.array([3])
 
@@ -57,7 +58,7 @@ def test_two_features():
 
 
 def test_multi_class():
-    estimator = linear_model.LogisticRegression()
+    estimator = LogisticRegression()
     estimator.coef_ = np.array([[1, 2], [3, 4], [5, 6]])
     estimator.intercept_ = np.array([7, 8, 9])
 
@@ -109,7 +110,7 @@ def test_multi_class():
 
 
 def test_binary_class():
-    estimator = linear_model.LogisticRegression()
+    estimator = LogisticRegression()
     estimator.coef_ = np.array([[1, 2]])
     estimator.intercept_ = np.array([3])
 
@@ -135,7 +136,7 @@ def test_binary_class():
 
 def test_ransac_custom_base_estimator():
     base_estimator = DecisionTreeRegressor()
-    estimator = linear_model.RANSACRegressor(
+    estimator = RANSACRegressor(
         base_estimator=base_estimator,
         random_state=1)
     estimator.fit([[1], [2], [3]], [1, 2, 3])
@@ -157,7 +158,7 @@ def test_ransac_custom_base_estimator():
 @pytest.mark.xfail(raises=NotImplementedError, strict=True)
 def test_ransac_unknown_base_estimator():
     base_estimator = DummyRegressor()
-    estimator = linear_model.RANSACRegressor(
+    estimator = RANSACRegressor(
         base_estimator=base_estimator,
         random_state=1)
     estimator.fit([[1], [2], [3]], [1, 2, 3])

tests/assemblers/test_svm.py

@@ -1,13 +1,13 @@
 import pytest
 import numpy as np
-from sklearn import svm
+from sklearn.svm import SVC
 from lightning.classification import KernelSVC
 from m2cgen import assemblers, ast
 from tests import utils
 
 
 def test_rbf_kernel():
-    estimator = svm.SVC(kernel="rbf", random_state=1, gamma=2.0)
+    estimator = SVC(kernel="rbf", random_state=1, gamma=2.0)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -22,7 +22,7 @@ def test_rbf_kernel():
 
 
 def test_linear_kernel():
-    estimator = svm.SVC(kernel="linear", random_state=1)
+    estimator = SVC(kernel="linear", random_state=1)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -43,7 +43,7 @@ def kernel_ast(sup_vec_value):
 
 
 def test_sigmoid_kernel():
-    estimator = svm.SVC(kernel="sigmoid", random_state=1, gamma=2.0)
+    estimator = SVC(kernel="sigmoid", random_state=1, gamma=2.0)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -71,7 +71,7 @@ def kernel_ast(sup_vec_value):
 
 
 def test_poly_kernel():
-    estimator = svm.SVC(kernel="poly", random_state=1, gamma=2.0, degree=2)
+    estimator = SVC(kernel="poly", random_state=1, gamma=2.0, degree=2)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -137,7 +137,7 @@ def kernel_ast(sup_vec_value):
 
 @pytest.mark.xfail(raises=ValueError, strict=True)
 def test_unknown_kernel():
-    estimator = svm.SVC(kernel=lambda x, y: np.transpose(x) * y)
+    estimator = SVC(kernel=lambda x, y: np.transpose(x) * y)
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -146,7 +146,7 @@ def test_unknown_kernel():
 
 
 def test_multi_class_rbf_kernel():
-    estimator = svm.SVC(kernel="rbf", random_state=1, gamma=2.0)
+    estimator = SVC(kernel="rbf", random_state=1, gamma=2.0)
 
     estimator.fit([[1], [2], [3]], [1, 2, 3])
 

tests/assemblers/test_tree.py

@@ -1,11 +1,11 @@
-from sklearn import tree
+from sklearn.tree import DecisionTreeRegressor, DecisionTreeClassifier
 
 from m2cgen import assemblers, ast
 from tests import utils
 
 
 def test_single_condition():
-    estimator = tree.DecisionTreeRegressor()
+    estimator = DecisionTreeRegressor()
 
     estimator.fit([[1], [2]], [1, 2])
 
@@ -24,7 +24,7 @@ def test_single_condition():
 
 
 def test_two_conditions():
-    estimator = tree.DecisionTreeRegressor()
+    estimator = DecisionTreeRegressor()
 
     estimator.fit([[1], [2], [3]], [1, 2, 3])
 
@@ -49,7 +49,7 @@ def test_two_conditions():
 
 
 def test_multi_class():
-    estimator = tree.DecisionTreeClassifier()
+    estimator = DecisionTreeClassifier()
 
     estimator.fit([[1], [2], [3]], [0, 1, 2])
 

tests/e2e/test_cli.py

@@ -3,7 +3,7 @@
 import platform
 import subprocess
 
-from sklearn import linear_model
+from sklearn.linear_model import LinearRegression
 
 from tests import utils
 
@@ -22,7 +22,7 @@ def execute_test(exec_args):
 def _prepare_pickled_model(tmp_path):
     p = tmp_path / "model.pickle"
 
-    estimator = linear_model.LinearRegression()
+    estimator = LinearRegression()
     utils.get_regression_model_trainer()(estimator)
 
     p.write_bytes(pickle.dumps(estimator))