update preprocess function to accommodate single row scoring

Author: djm21

examples/pzmm_generate_complete_model_card.ipynb

@@ -567,7 +567,11 @@
     "    df = pd.get_dummies(df, columns=[\"WorkClass\", \"Education\", \"MartialStatus\", \"Relationship\", \"Race\", \"Sex\"])\n",
     "    df.columns = df.columns.str.replace(' ', '')\n",
     "    df.columns = df.columns.str.replace('-', '_')\n",
-    "    df = df.drop(['Sex_Male'], axis=1)\n",
+    "    # Ensure that Sex_Female column exists and Sex_Male column is dropped if it exists\n",
+    "    if 'Sex_Female' not in df.columns:\n",
+    "        df['Sex_Female'] = 0\n",
+    "    if 'Sex_Male' in df.columns:\n",
+    "        df = df.drop(['Sex_Male'], axis=1)\n",
     "    if 'index' in df.columns or 'index' in cat_vals.columns:\n",
     "        df = pd.concat([df, cat_vals], axis=1).drop('index', axis=1)\n",
     "    # For the model to score correctly, all OHE columns must exist\n",