josedev_1.0_20230223_1128

Author: JNevado81

test/test_export_model.py

@@ -0,0 +1,17 @@
+from toolkit.machine_learning import export_model
+
+from sklearn.neighbors import KNeighborsClassifier
+import os
+
+import pytest
+
+def test_export_model():
+    model = KNeighborsClassifier()
+    dir_model = os.path.join(os.getcwd(), 'model')
+    # dir_model = os.getcwd() + '\\model'
+    # # Guardamos en una variable el directorio del script
+    # dir_script = os.path.dirname(os.path.abspath(__file__))
+    # # Guardamos la ubicacion del modelo
+    # dir_model = os.path.join(dir_script, 'model')
+
+    assert export_model(model, dir_model=dir_model, name_model='model_the_best') == None

test/test_import_model.py

@@ -0,0 +1,17 @@
+from toolkit.machine_learning import import_model
+
+from sklearn.neighbors import KNeighborsClassifier
+import os
+
+import pytest
+
+def test_import_model():
+    model_import = KNeighborsClassifier()
+    dir_model = os.path.join(os.getcwd(), 'model_KNN')
+    # dir_model = os.getcwd() + '\\model_KNN'
+    # # Guardamos en una variable el directorio del script
+    # dir_script = os.path.dirname(os.path.abspath(__file__))
+    # # Guardamos la ubicacion del modelo
+    # dir_model = os.path.join(dir_script, 'model_KNN')
+
+    assert import_model(dir_model=dir_model, name_model='model_KNN') == model_import

test/test_predict_model_classification.py

@@ -0,0 +1,22 @@
+from toolkit.machine_learning import predict_model_classification
+
+from sklearn.neighbors import KNeighborsClassifier
+import pandas as pd
+
+import pytest
+
+def test_predict_model_classification():
+    model_import = KNeighborsClassifier()
+    X_test = pd.DataFrame({'sex':[0, 1], 
+                   'hypertension': [1, 0],
+                   'heart_disease':[1, 0],
+                   'Residence_type':[0, 1]}
+                   )
+    y_test = pd.DataFrame({'':[1, 1]})
+    y_pred_test = [0, 0] # [0 0]
+
+    assert predict_model_classification(model_import, X_test, y_test, test_score=False) == X_test, y_pred_test
+
+    # X_test_r, y_pred_test_r = predict_model_classification(model_import, X_test, y_test, test_score=False)
+    # assert X_test_r == X_test
+    # assert y_pred_test_r.tolist() == y_pred_test

test/test_processing_model_classification.py

@@ -0,0 +1,49 @@
+from toolkit.machine_learning import processing_model_classification
+
+from sklearn.neighbors import KNeighborsClassifier
+import pandas as pd
+
+import pytest
+
+def test_processing_model_classification():
+    model = KNeighborsClassifier()
+    df = pd.DataFrame({'sex': [0, 1, 1, 0, 1, 1, 0], 'hypertension': [0, 0, 1, 1, 0, 0, 1], 'heart_disease': [0, 0, 1, 1, 0, 1, 1], 'Residence_type': [1, 1, 0, 1, 0, 1, 0], 'stroke': [0, 1, 1, 0, 1, 0, 1]})
+    X = df.drop(columns=['stroke'])
+    y = df['stroke']
+
+    X_train = pd.DataFrame({'sex':[0, 1, 1, 1, 0], 
+                   'hypertension': [0, 1, 0, 0, 1],
+                   'heart_disease':[0, 1, 0, 1, 1],
+                   'Residence_type':[1, 0, 0, 1, 1]}
+                   )
+    X_test = pd.DataFrame({'sex':[0, 1], 
+                   'hypertension': [1, 0],
+                   'heart_disease':[1, 0],
+                   'Residence_type':[0, 1]}
+                   )
+    y_train = pd.DataFrame({'':[0, 1, 1, 0, 0]})
+    y_test = pd.DataFrame({'':[1, 1]})
+    y_pred_train = [0, 0, 0, 0, 0] # [0 0 0 0 0]
+    y_pred_test = [0, 0] # [0 0]
+
+    model_r, X_train_r, X_test_r, y_train_r, y_test_r, y_pred_train_r, y_pred_test_r = processing_model_classification(model, X, y, test_size_split=0.20, shuffle_split=True, 
+                                                                                                         random_state_split=32, standardScaler=False, train_score=True, test_score=False)
+
+    assert model_r == model
+    assert X_train_r == X_train
+    assert X_test_r == X_test
+    assert y_train_r == y_train
+    assert y_test_r == y_test
+    assert y_pred_train_r == y_pred_train
+    assert y_pred_test_r == y_pred_test
+
+    # assert isinstance(model_r, model)
+    # assert isinstance(X_train_r, X_train)
+    # assert isinstance(X_test_r, X_test)
+    # assert isinstance(y_train_r, y_train)
+    # assert isinstance(y_test_r, y_test)
+    # assert isinstance(y_pred_train_r, y_pred_train)
+    # assert isinstance(y_pred_test_r, y_pred_test)
+
+    # assert processing_model_classification(model, X, y, test_size_split=0.20, shuffle_split=True, standardScaler=False, 
+    #                                        train_score=True, test_score=False) == model, X_train, X_test, y_train, y_test, y_pred_train, y_pred_test

test/test_show_scoring.py

@@ -0,0 +1,15 @@
+from toolkit.machine_learning import show_scoring
+import pandas as pd
+
+import pytest
+
+def test_show_scoring():
+    y_test = pd.DataFrame({'':[1, 1]})
+    y_pred_test = [0, 0] # [0 0]
+
+    dict_scoring = show_scoring(y_test, y_pred_test, label='TEST', round=1, auc_sc=False, roc_auc_sc=False, confusion_matrix_sc=False)
+
+    assert dict_scoring['accu_r'] == 0.0
+    assert dict_scoring['auc_r'] == None
+    assert dict_scoring['roc_auc_r'] == None
+    assert dict_scoring['conf_mat_r'] == None