Refactor the code

Author: scorebot

.github/workflows/python-publish.yml

@@ -30,6 +30,10 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install build
+    - name: Test with pytest
+      run: |
+        pip install pytest pytest-cov
+        pytest
     - name: Build package
       run: python -m build --sdist
     - name: Publish package

pypmml/jvm.py

@@ -26,8 +26,7 @@ def __init__(self):
         self.java_path = None
 
         # JVM options
-        # Fix IllegalAccessError: cannot access class jdk.internal.math.FloatingDecimal
-        self.java_opts = ["--add-exports=java.base/jdk.internal.math=ALL-UNNAMED"]
+        self.java_opts = []
         java_opts = os.environ.get("JAVA_OPTS")
         if java_opts:
             self.java_opts.extend(java_opts.split())
@@ -144,7 +143,7 @@ def name(self):
 
 class Py4jGateway(JVMGateway):
     """Py4j"""
-    from py4j.java_collections import JavaArray
+    from py4j.java_collections import JavaArray, JavaList
     from py4j.java_gateway import JavaObject
     from py4j.protocol import Py4JJavaError
 
@@ -179,7 +178,7 @@ def detach(self, java_object):
         Py4jGateway._gateway.detach(java_object)
 
     def java2py(self, r):
-        if isinstance(r, self.JavaArray):
+        if isinstance(r, (self.JavaArray, self.JavaList)):
             return [self.java2py(x) for x in r]
         elif isinstance(r, self.JavaObject):
             cls_name = r.getClass().getName()

pypmml/metadata.py

@@ -58,7 +58,11 @@ def targetField(self):
 
     @property
     def value(self):
-        return self.call('value')
+        obj = self.call('value')
+        if obj is not None:
+            return DataVal(obj).toVal
+        else:
+            return None
 
     @property
     def ruleFeature(self):
@@ -88,3 +92,12 @@ def fieldNames(self):
     def get(self, name):
         fld = self.get(name)
         return Field(fld) if fld is not None else None
+
+
+class DataVal(JavaModelWrapper):
+    def __init__(self, java_model):
+        super(DataVal, self).__init__(java_model)
+
+    @property
+    def toVal(self):
+        return self.call('toVal')

pypmml/model.py

@@ -19,7 +19,8 @@
 from pypmml.base import JavaModelWrapper, PMMLContext
 from pypmml.jvm import PMMLError
 from pypmml.elements import Header
-from pypmml.metadata import Field, OutputField, DataDictionary
+from pypmml.metadata import Field, OutputField, DataDictionary, DataVal
+from pypmml.utils import is_nd_array, is_pandas_series, is_pandas_dataframe
 
 
 class Model(JavaModelWrapper):
@@ -115,33 +116,13 @@ def outputFields(self):
     @property
     def classes(self):
         """The class labels in a classification model."""
-        return self.call('classes')
+        values = self.call('classes')
+        return [DataVal(x).toVal for x in values]
 
     def setSupplementOutput(self, value):
         self.call('setSupplementOutput', value)
         return self
 
-    def _is_nd_array(self, data):
-        try:
-            import numpy as np
-            return isinstance(data, np.ndarray)
-        except ImportError:
-            return False
-
-    def _is_pandas_dataframe(self, data):
-        try:
-            import pandas as pd
-            return isinstance(data, pd.DataFrame)
-        except ImportError:
-            return False
-
-    def _is_pandas_series(self, data):
-        try:
-            import pandas as pd
-            return isinstance(data, pd.Series)
-        except ImportError:
-            return False
-
     def predict(self, data):
         """
         Predict values for a given data.
@@ -163,8 +144,7 @@ def predict(self, data):
                         return self.call('predict', data)
                 else:
                     return []
-            elif self._is_nd_array(data):
-                import numpy as np
+            elif is_nd_array(data):
                 if data.ndim == 1:
                     return self.call('predict', data.tolist())
                 elif data.ndim == 2:
@@ -179,13 +159,13 @@ def predict(self, data):
                         return [self.call('predict', record.tolist()) for record in data]
                 else:
                     raise PMMLError('Max 2 dimensions are supported')
-            elif self._is_pandas_dataframe(data):
+            elif is_pandas_dataframe(data):
                 import pandas as pd
                 from io import StringIO
                 json_data = data.to_json(orient='split', index=False)
                 result = self.call('predict', json_data)
                 return pd.read_json(StringIO(result), orient='split')
-            elif self._is_pandas_series(data):
+            elif is_pandas_series(data):
                 import pandas as pd
                 record = data.to_dict()
                 result = self.call('predict', record)

pypmml/test/test_model.py

@@ -16,15 +16,19 @@
 
 import unittest
 from unittest import TestCase
+from os import path
 
 from pypmml import Model, PMMLContext
 
 
 class ModelTestCase(TestCase):
+    test_data_dir = path.join(path.dirname(__file__), 'resources', 'data')
+    test_models_dir = path.join(path.dirname(__file__), 'resources', 'models')
 
     def test_from_file(self):
         # The model is from here: http://dmg.org/pmml/pmml_examples/KNIME_PMML_4.1_Examples/single_iris_dectree.xml
-        model = Model.fromFile('./resources/models/single_iris_dectree.xml')
+        model_path = path.join(self.test_models_dir, 'single_iris_dectree.xml')
+        model = Model.fromFile(model_path)
         self.assertEqual(model.version, '4.1')
 
         app = model.header.application
@@ -115,7 +119,8 @@ def test_from_file(self):
     def test_pandas(self):
         try:
             import pandas as pd
-            model = Model.load('./resources/models/single_iris_dectree.xml')
+            model_path = path.join(self.test_models_dir, 'single_iris_dectree.xml')
+            model = Model.load(model_path)
 
             # Data in Series
             result = model.predict(pd.Series({'sepal_length': 5.1, 'sepal_width': 3.5, 'petal_length': 1.4, 'petal_width': 0.2}))
@@ -124,7 +129,8 @@ def test_pandas(self):
             self.assertEqual(result.get('node_id'), '1')
 
             # Data in DataFrame
-            data = pd.read_csv('./resources/data/Iris.csv')
+            data_path = path.join(self.test_data_dir, 'Iris.csv')
+            data = pd.read_csv(data_path)
             result = model.predict(data)
             self.assertEqual(result.iloc[0].get('predicted_class'), 'Iris-setosa')
             self.assertEqual(result.iloc[0].get('probability'), 1.0)
@@ -135,7 +141,8 @@ def test_pandas(self):
     def test_numpy(self):
         try:
             import numpy as np
-            model = Model.load('./resources/models/single_iris_dectree.xml')
+            model_path = path.join(self.test_models_dir, 'single_iris_dectree.xml')
+            model = Model.load(model_path)
 
             # Data in 1-D
             result = model.predict(np.array([5.1, 3.5, 1.4, 0.2]))
@@ -166,7 +173,7 @@ def test_numpy(self):
             pass
 
     def test_load(self):
-        file_path = './resources/models/single_iris_dectree.xml'
+        file_path = path.join(self.test_models_dir, 'single_iris_dectree.xml')
         self.assertTrue(Model.load(file_path) is not None)
 
         with open(file_path, 'rb') as f:

pypmml/utils.py

@@ -0,0 +1,38 @@
+#
+# Copyright (c) 2024 AutoDeployAI
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+def is_nd_array(data):
+    try:
+        import numpy as np
+        return isinstance(data, np.ndarray)
+    except ImportError:
+        return False
+
+
+def is_pandas_dataframe(data):
+    try:
+        import pandas as pd
+        return isinstance(data, pd.DataFrame)
+    except ImportError:
+        return False
+
+
+def is_pandas_series(data):
+    try:
+        import pandas as pd
+        return isinstance(data, pd.Series)
+    except ImportError:
+        return False

pypmml/version.py

@@ -18,4 +18,4 @@
 
 # Dev branch marker is: 'X.Y.dev' or 'X.Y.devN' where N is an integer.
 # 'X.Y.dev0' is the canonical version of 'X.Y.dev'
-__version__ = '1.5.2'
+__version__ = '1.5.3'