Upload P2 - Secom Notebook - Mercury.ipynb

P2 - Secom Notebook - Mercury.ipynb

@@ -15,18 +15,14 @@
   {
    "attachments": {},
    "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "skip"
-    }
-   },
+   "metadata": {},
    "source": [
     "#### **Import the data**"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 97,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -53,7 +49,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 98,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -64,7 +60,7 @@
      "data": {
       "application/mercury+json": {
        "allow_download": true,
-       "code_uid": "App.0.40.24.1-rand0e93859a",
+       "code_uid": "App.0.40.24.1-rand0254d4eb",
        "continuous_update": false,
        "description": "Recumpute everything dynamically",
        "full_screen": true,
@@ -96,7 +92,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 99,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -138,7 +134,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
+   "execution_count": 100,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -195,7 +191,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
+   "execution_count": 101,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -290,7 +286,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
+   "execution_count": 102,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -341,7 +337,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 48,
+   "execution_count": 103,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -419,7 +415,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 104,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -485,9 +481,21 @@
     "        return df_imputed\n"
    ]
   },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {
+    "slideshow": {
+     "slide_type": "skip"
+    }
+   },
+   "source": [
+    "#### **Feature Selection**"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 105,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -507,6 +515,7 @@
     "        selected_columns = X_train.columns[selected_feature_indices]\n",
     "        X_train_filtered = X_train.iloc[:, selected_feature_indices]\n",
     "        print('Shape of the training set after feature selection with Boruta: ', X_train_filtered.shape)\n",
+    "        feature_selection_var = 'boruta'\n",
     "        return X_train_filtered, selected_columns\n",
     "    \n",
     "    if method == 'none':\n",
@@ -515,6 +524,7 @@
     "        print('Shape of the training set after no feature selection: ', X_train_filtered.shape)\n",
     "        feature_selection_var = 'none'\n",
     "        selected_features = X_train_filtered.columns\n",
+    "        feature_selection_var = 'none'\n",
     "        return X_train_filtered, selected_features        \n",
     "    \n",
     "    if method == 'lasso':\n",
@@ -567,7 +577,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 106,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -585,8 +595,6 @@
     "        from imblearn.over_sampling import SMOTE\n",
     "        sm = SMOTE(random_state=42)\n",
     "        X_train_res, y_train_res = sm.fit_resample(X_train, y_train)\n",
-    "        print('Shape of the training set after oversampling with SMOTE: ', X_train_res.shape)\n",
-    "        print('Value counts of the target variable after oversampling with SMOTE: \\n', y_train_res.value_counts())\n",
     "        imbalance_var = 'smote'\n",
     "        return X_train_res, y_train_res\n",
     "    \n",
@@ -594,8 +602,6 @@
     "        from imblearn.under_sampling import RandomUnderSampler\n",
     "        rus = RandomUnderSampler(random_state=42)\n",
     "        X_train_res, y_train_res = rus.fit_resample(X_train, y_train)\n",
-    "        print('Shape of the training set after undersampling with RandomUnderSampler: ', X_train_res.shape)\n",
-    "        print('Value counts of the target variable after undersampling with RandomUnderSampler: \\n', y_train_res.value_counts())\n",
     "        imbalance_var = 'random_undersampling'\n",
     "        return X_train_res, y_train_res\n",
     "    \n",
@@ -603,8 +609,6 @@
     "        from imblearn.over_sampling import RandomOverSampler\n",
     "        ros = RandomOverSampler(random_state=42)\n",
     "        X_train_res, y_train_res = ros.fit_resample(X_train, y_train)\n",
-    "        print('Shape of the training set after oversampling with RandomOverSampler: ', X_train_res.shape)\n",
-    "        print('Value counts of the target variable after oversampling with RandomOverSampler: \\n', y_train_res.value_counts())\n",
     "        imbalance_var = 'rose'\n",
     "        return X_train_res, y_train_res\n",
     "    \n",
@@ -612,8 +616,6 @@
     "    if method == 'none':\n",
     "        X_train_res = X_train\n",
     "        y_train_res = y_train\n",
-    "        print('Shape of the training set after no resampling: ', X_train_res.shape)\n",
-    "        print('Value counts of the target variable after no resampling: \\n', y_train_res.value_counts())\n",
     "        imbalance_var = 'none'\n",
     "        return X_train_res, y_train_res\n",
     "    \n",
@@ -638,7 +640,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 52,
+   "execution_count": 107,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -727,7 +729,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
+   "execution_count": 108,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -741,13 +743,12 @@
     "\n",
     "def evaluate_models(model='random_forest'):\n",
     "    print('Have the duplicates been removed?', drop_duplicates_var)\n",
-    "    print('What is the missing values threshold %?', missing_values_threshold_var)\n",
-    "    print('What is the variance threshold?', variance_threshold_var)\n",
-    "    print('What is the correlation threshold?', correlation_threshold_var)\n",
-    "    print('What is the outlier removal threshold?', outlier_var)\n",
-    "    print('What is the scaling method?', scale_model_var)\n",
-    "    print('What is the imputation method?', imputation_var) \n",
-    "    print('What is the imbalance treatment?', imbalance_var)\n",
+    "    print('Missing values threshold is:', missing_values_threshold_var,' - Variance threshold is:,', variance_threshold_var,' - Correlation threshold is:', correlation_threshold_var)\n",
+    "    print('Outlier removal threshold is:', outlier_var)\n",
+    "    print('Scaling method is:', scale_model_var)\n",
+    "    print('Imputation method is:', imputation_var)\n",
+    "    print('Feature selection method is:', feature_selection_var)\n",
+    "    print('Imbalance treatment method is:', imbalance_var)\n",
     "\n",
     "    all_models = ['random_forest', 'logistic_regression', 'knn', 'svm', 'naive_bayes', 'decision_tree', 'xgboost']\n",
     "    evaluation_score_append = []\n",
@@ -816,7 +817,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 54,
+   "execution_count": 113,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -830,17 +831,17 @@
         "yes",
         "no"
        ],
-       "code_uid": "Select.0.40.16.25-rand98b210b9",
+       "code_uid": "Select.0.40.16.25-rand2694baac",
        "disabled": false,
        "hidden": false,
        "label": "Drop Duplicates",
-       "model_id": "f2d0fb31478a477ea1b0d4c4aa80fb2e",
+       "model_id": "d2be0f72c8ad4556970977c13a04a1c8",
        "url_key": "",
        "value": "yes",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "f2d0fb31478a477ea1b0d4c4aa80fb2e",
+       "model_id": "d2be0f72c8ad4556970977c13a04a1c8",
        "version_major": 2,
        "version_minor": 0
       },
@@ -854,18 +855,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.28-randc77c765e",
+       "code_uid": "Text.0.40.15.28-rande1621478",
        "disabled": false,
        "hidden": false,
        "label": "Missing Value Threeshold",
-       "model_id": "f97f46b880e3434082498e1e720b0b65",
+       "model_id": "80b2d08ffcd84257b8bc791ed6e20d5a",
        "rows": 1,
        "url_key": "",
        "value": "80",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "f97f46b880e3434082498e1e720b0b65",
+       "model_id": "80b2d08ffcd84257b8bc791ed6e20d5a",
        "version_major": 2,
        "version_minor": 0
       },
@@ -879,18 +880,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.31-rand54dd8817",
+       "code_uid": "Text.0.40.15.31-rand4541ac63",
        "disabled": false,
        "hidden": false,
        "label": "Variance Threshold",
-       "model_id": "4839b45f7e9b483d9fb8ee7fc05f1e19",
+       "model_id": "be3aff94bc9946d283b7b34af9b61b1d",
        "rows": 1,
        "url_key": "",
        "value": "0",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4839b45f7e9b483d9fb8ee7fc05f1e19",
+       "model_id": "be3aff94bc9946d283b7b34af9b61b1d",
        "version_major": 2,
        "version_minor": 0
       },
@@ -904,18 +905,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.34-rand811824bd",
+       "code_uid": "Text.0.40.15.34-rand92107f0e",
        "disabled": false,
        "hidden": false,
        "label": "Correlation Threshold",
-       "model_id": "10033d424ab949f7b51462e444e17ba7",
+       "model_id": "20a2b896579147a4a8cf1f8593ca263c",
        "rows": 1,
        "url_key": "",
        "value": "1",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "10033d424ab949f7b51462e444e17ba7",
+       "model_id": "20a2b896579147a4a8cf1f8593ca263c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -935,17 +936,17 @@
         4,
         5
        ],
-       "code_uid": "Select.0.40.16.38-rand10d00d99",
+       "code_uid": "Select.0.40.16.38-randf6cb87b9",
        "disabled": false,
        "hidden": false,
        "label": "Outlier Removal Threshold",
-       "model_id": "96b8980bceaf46459d9ec06c8fb7c818",
+       "model_id": "013f6eeb57534a49abd399f13c4814aa",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "96b8980bceaf46459d9ec06c8fb7c818",
+       "model_id": "013f6eeb57534a49abd399f13c4814aa",
        "version_major": 2,
        "version_minor": 0
       },
@@ -966,17 +967,17 @@
         "minmax",
         "robust"
        ],
-       "code_uid": "Select.0.40.16.46-rand1bc79c9d",
+       "code_uid": "Select.0.40.16.46-rand035a3a64",
        "disabled": false,
        "hidden": false,
        "label": "Scaling Variables",
-       "model_id": "e7650cea7a834d588a995407052e1f2c",
+       "model_id": "4c799a2109f0475dad281094174aff03",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e7650cea7a834d588a995407052e1f2c",
+       "model_id": "4c799a2109f0475dad281094174aff03",
        "version_major": 2,
        "version_minor": 0
       },
@@ -996,17 +997,17 @@
         "knn",
         "most_frequent"
        ],
-       "code_uid": "Select.0.40.16.50-rand69ae31a0",
+       "code_uid": "Select.0.40.16.50-rand1c821039",
        "disabled": false,
        "hidden": false,
        "label": "Imputation Methods",
-       "model_id": "652a64af16174970919183e6ab1c5b53",
+       "model_id": "7c5bf031dc55488688a5878edb7cc55f",
        "url_key": "",
        "value": "mean",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "652a64af16174970919183e6ab1c5b53",
+       "model_id": "7c5bf031dc55488688a5878edb7cc55f",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1027,17 +1028,17 @@
         "pca",
         "boruta"
        ],
-       "code_uid": "Select.0.40.16.55-rand148632f9",
+       "code_uid": "Select.0.40.16.55-randf440b52a",
        "disabled": false,
        "hidden": false,
        "label": "Feature Selection",
-       "model_id": "9a1a199471314cd7a3363ea25d9d341a",
+       "model_id": "f2b440a394e6473dbcf3ab959393d76f",
        "url_key": "",
-       "value": "none",
+       "value": "lasso",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9a1a199471314cd7a3363ea25d9d341a",
+       "model_id": "f2b440a394e6473dbcf3ab959393d76f",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1057,17 +1058,17 @@
         "undersampling",
         "rose"
        ],
-       "code_uid": "Select.0.40.16.59-rand3a34b3e3",
+       "code_uid": "Select.0.40.16.59-randd37e3f6b",
        "disabled": false,
        "hidden": false,
        "label": "Imbalance Treatment",
-       "model_id": "358dd80171af4de2a944c3077b2f48d8",
+       "model_id": "59021cfbc5d8465ba0dae8da4581ba65",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "358dd80171af4de2a944c3077b2f48d8",
+       "model_id": "59021cfbc5d8465ba0dae8da4581ba65",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1090,17 +1091,17 @@
         "decision_tree",
         "xgboost"
        ],
-       "code_uid": "Select.0.40.16.64-rand4b9cf5e0",
+       "code_uid": "Select.0.40.16.64-rand7458a327",
        "disabled": false,
        "hidden": false,
        "label": "Model Selection",
-       "model_id": "8477ca5211bd4914861b3e48cda21c10",
+       "model_id": "2e45c8c32a4e44a6b612ad6943d9890e",
        "url_key": "",
        "value": "random_forest",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "8477ca5211bd4914861b3e48cda21c10",
+       "model_id": "2e45c8c32a4e44a6b612ad6943d9890e",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1208,7 +1209,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 55,
+   "execution_count": 116,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -1234,12 +1235,8 @@
       "Number of missing values after imputation:  0\n",
       "Number of missing values before imputation:  6954\n",
       "Number of missing values after imputation:  0\n",
-      "Shape of the training set after no resampling:  (1175, 445)\n",
-      "Value counts of the target variable after no resampling: \n",
-      " pass/fail\n",
-      "0            1097\n",
-      "1              78\n",
-      "dtype: int64\n"
+      "Selected method is:  lasso\n",
+      "Shape of the training set after feature selection with LassoCV:  (1175, 6)\n"
      ]
     }
    ],
@@ -1266,9 +1263,16 @@
     "X_train_imputed = impute_missing_values(input_imputation_method,X_train_scaled, X_train_scaled, input_n_neighbors)\n",
     "X_test_imputed = impute_missing_values(input_imputation_method,X_train_scaled, X_test_scaled, input_n_neighbors)\n",
     "\n",
+    "# select the features using the function feature_selection\n",
+    "\n",
+    "X_train_selected, selected_features = feature_selection(input_feature_selection, X_train_imputed, y_train)\n",
+    "\n",
+    "X_train_selected = pd.DataFrame(X_train_selected, columns=selected_features)\n",
+    "X_test_selected = X_test_imputed[selected_features]\n",
+    "\n",
     "# treat imbalance in the training set using the function oversample\n",
     "\n",
-    "X_train_res, y_train_res = imbalance_treatment(input_imbalance_treatment, X_train_imputed, y_train)\n",
+    "X_train_res, y_train_res = imbalance_treatment(input_imbalance_treatment, X_train_selected, y_train)\n",
     "\n"
    ]
   },
@@ -1286,7 +1290,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 56,
+   "execution_count": null,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -1324,7 +1328,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 57,
+   "execution_count": null,
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
@@ -1336,125 +1340,23 @@
      "output_type": "stream",
      "text": [
       "Have the duplicates been removed? yes\n",
-      "What is the missing values threshold %? 80\n",
-      "What is the variance threshold? 0.0\n",
-      "What is the correlation threshold? 1.0\n",
-      "What is the outlier removal threshold? none\n",
-      "What is the scaling method? none\n",
-      "What is the imputation method? mean\n",
-      "What is the imbalance treatment? none\n"
+      "Missing values threshold is: 80  - Variance threshold is:, 0.0  - Correlation threshold is: 1.0\n",
+      "Outlier removal threshold is: none\n",
+      "Scaling method is: none\n",
+      "Imputation method is: mean\n"
      ]
     },
     {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>Model</th>\n",
-       "      <th>Accuracy</th>\n",
-       "      <th>Precision</th>\n",
-       "      <th>Recall</th>\n",
-       "      <th>F1-score</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>random_forest</td>\n",
-       "      <td>0.93</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "           Model  Accuracy  Precision  Recall  F1-score\n",
-       "0  random_forest      0.93        0.0     0.0       0.0"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>Model</th>\n",
-       "      <th>True Negatives</th>\n",
-       "      <th>False Positives</th>\n",
-       "      <th>False Negatives</th>\n",
-       "      <th>True Positives</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>random_forest</td>\n",
-       "      <td>366</td>\n",
-       "      <td>0</td>\n",
-       "      <td>26</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "           Model  True Negatives  False Positives  False Negatives  \\\n",
-       "0  random_forest             366                0               26   \n",
-       "\n",
-       "   True Positives  \n",
-       "0               0  "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 350x350 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
+     "ename": "NameError",
+     "evalue": "name 'feature_selection_var' is not defined",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_34908\\804542050.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mevaluation_score_output\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mevaluation_counts_output\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mevaluate_models\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minput_model\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[1;31m# check if the model has already been evaluated and if not, append the results to the dataframe\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mevaluation_score_df\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mevaluation_score_output\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mevaluation_score_df\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mignore_index\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
+      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_34908\\293505729.py\u001b[0m in \u001b[0;36mevaluate_models\u001b[1;34m(model)\u001b[0m\n\u001b[0;32m      9\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Scaling method is:'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscale_model_var\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     10\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Imputation method is:'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mimputation_var\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 11\u001b[1;33m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Feature selection method is:'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeature_selection_var\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     12\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Imbalance treatment method is:'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mimbalance_var\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     13\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
+      "\u001b[1;31mNameError\u001b[0m: name 'feature_selection_var' is not defined"
+     ]
     }
    ],
    "source": [