diff --git a/Examples/KDD2009Example/KDD2009Example.ipynb b/Examples/KDD2009Example/KDD2009Example.ipynb
index 102f836..df494b8 100644
--- a/Examples/KDD2009Example/KDD2009Example.ipynb
+++ b/Examples/KDD2009Example/KDD2009Example.ipynb
@@ -50,7 +50,7 @@
     {
      "data": {
       "text/plain": [
-       "'0.4.4'"
+       "'0.5.0'"
       ]
      },
      "execution_count": 2,
@@ -469,8 +469,17 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "DataFrame.dtypes for data must be int, float or bool.\n",
-      "                Did not expect the data types in fields Var191, Var192, Var193, Var194, Var195, Var196, Var197, Var198, Var199, Var200, Var201, Var202, Var203, Var204, Var205, Var206, Var207, Var208, Var210, Var211, Var212, Var213, Var214, Var215, Var216, Var217, Var218, Var219, Var220, Var221, Var222, Var223, Var224, Var225, Var226, Var227, Var228, Var229\n"
+      "DataFrame.dtypes for data must be int, float, bool or categorical.  When\n",
+      "                categorical type is supplied, DMatrix parameter\n",
+      "                `enable_categorical` must be set to `True`.Var191, Var192, Var193, Var194, Var195, Var196, Var197, Var198, Var199, Var200, Var201, Var202, Var203, Var204, Var205, Var206, Var207, Var208, Var210, Var211, Var212, Var213, Var214, Var215, Var216, Var217, Var218, Var219, Var220, Var221, Var222, Var223, Var224, Var225, Var226, Var227, Var228, Var229\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/xgboost/sklearn.py:888: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
+      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n"
      ]
     }
    ],
@@ -508,7 +517,10 @@
    "source": [
     "plan = vtreat.BinomialOutcomeTreatment(\n",
     "    outcome_target=True,\n",
-    "    params=vtreat.vtreat_parameters({'filter_to_recommended':True}))"
+    "    params=vtreat.vtreat_parameters({\n",
+    "        'filter_to_recommended': True,\n",
+    "        'sparse_indicators': True,\n",
+    "    }))"
    ]
   },
   {
@@ -568,27 +580,27 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>Var2_is_bad</th>\n",
-       "      <th>Var3_is_bad</th>\n",
-       "      <th>Var4_is_bad</th>\n",
-       "      <th>Var5_is_bad</th>\n",
-       "      <th>Var6_is_bad</th>\n",
-       "      <th>Var7_is_bad</th>\n",
-       "      <th>Var10_is_bad</th>\n",
        "      <th>Var11_is_bad</th>\n",
-       "      <th>Var13_is_bad</th>\n",
-       "      <th>Var14_is_bad</th>\n",
+       "      <th>Var82_is_bad</th>\n",
+       "      <th>Var144_is_bad</th>\n",
+       "      <th>Var27_is_bad</th>\n",
+       "      <th>Var54_is_bad</th>\n",
+       "      <th>Var107_is_bad</th>\n",
+       "      <th>Var19_is_bad</th>\n",
+       "      <th>Var74_is_bad</th>\n",
+       "      <th>Var117_is_bad</th>\n",
+       "      <th>Var159_is_bad</th>\n",
        "      <th>...</th>\n",
-       "      <th>Var227_lev_RAYp</th>\n",
-       "      <th>Var227_lev_ZI9m</th>\n",
-       "      <th>Var228_logit_code</th>\n",
-       "      <th>Var228_prevalence_code</th>\n",
-       "      <th>Var228_lev_F2FyR07IdsN7I</th>\n",
-       "      <th>Var229_logit_code</th>\n",
-       "      <th>Var229_prevalence_code</th>\n",
-       "      <th>Var229_lev__NA_</th>\n",
-       "      <th>Var229_lev_am7c</th>\n",
-       "      <th>Var229_lev_mj86</th>\n",
+       "      <th>Var191_lev__NA_</th>\n",
+       "      <th>Var191_lev_r__I</th>\n",
+       "      <th>Var213_logit_code</th>\n",
+       "      <th>Var213_prevalence_code</th>\n",
+       "      <th>Var213_lev__NA_</th>\n",
+       "      <th>Var213_lev_KdSa</th>\n",
+       "      <th>Var201_logit_code</th>\n",
+       "      <th>Var201_prevalence_code</th>\n",
+       "      <th>Var201_lev__NA_</th>\n",
+       "      <th>Var201_lev_smXZ</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
@@ -596,95 +608,95 @@
        "      <th>0</th>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>...</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
-       "      <td>0.145563</td>\n",
-       "      <td>0.654178</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>0.180634</td>\n",
-       "      <td>0.568733</td>\n",
+       "      <td>0.006417</td>\n",
+       "      <td>0.977733</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
+       "      <td>0.036642</td>\n",
+       "      <td>0.744956</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>...</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
-       "      <td>0.150727</td>\n",
-       "      <td>0.654178</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>0.175825</td>\n",
-       "      <td>0.568733</td>\n",
+       "      <td>0.008288</td>\n",
+       "      <td>0.977733</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
+       "      <td>0.039494</td>\n",
+       "      <td>0.744956</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>...</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
+       "      <td>0.008288</td>\n",
+       "      <td>0.977733</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
-       "      <td>-0.591072</td>\n",
-       "      <td>0.053667</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>-0.296854</td>\n",
-       "      <td>0.233689</td>\n",
+       "      <td>-0.125971</td>\n",
+       "      <td>0.254956</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>...</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
-       "      <td>0.150727</td>\n",
-       "      <td>0.654178</td>\n",
+       "      <td>0.008288</td>\n",
+       "      <td>0.977733</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>-0.292587</td>\n",
-       "      <td>0.196044</td>\n",
        "      <td>0.0</td>\n",
+       "      <td>-0.125971</td>\n",
+       "      <td>0.254956</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "    </tr>\n",
@@ -692,68 +704,68 @@
        "      <th>4</th>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>...</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
+       "      <td>0.008026</td>\n",
+       "      <td>0.977733</td>\n",
+       "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
-       "      <td>-0.323715</td>\n",
-       "      <td>0.018556</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>-0.268261</td>\n",
-       "      <td>0.233689</td>\n",
-       "      <td>0.0</td>\n",
+       "      <td>0.036528</td>\n",
+       "      <td>0.744956</td>\n",
        "      <td>1.0</td>\n",
        "      <td>0.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>5 rows × 233 columns</p>\n",
+       "<p>5 rows × 261 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "   Var2_is_bad  Var3_is_bad  Var4_is_bad  Var5_is_bad  Var6_is_bad  \\\n",
-       "0          1.0          1.0          1.0          1.0          0.0   \n",
-       "1          1.0          1.0          1.0          1.0          0.0   \n",
-       "2          1.0          1.0          1.0          1.0          0.0   \n",
-       "3          1.0          1.0          1.0          1.0          0.0   \n",
-       "4          1.0          1.0          1.0          1.0          0.0   \n",
+       "   Var11_is_bad  Var82_is_bad  Var144_is_bad  Var27_is_bad  Var54_is_bad  \\\n",
+       "0           1.0           1.0            0.0           1.0           1.0   \n",
+       "1           1.0           1.0            0.0           1.0           1.0   \n",
+       "2           1.0           1.0            0.0           1.0           1.0   \n",
+       "3           1.0           1.0            0.0           1.0           1.0   \n",
+       "4           1.0           1.0            0.0           1.0           1.0   \n",
        "\n",
-       "   Var7_is_bad  Var10_is_bad  Var11_is_bad  Var13_is_bad  Var14_is_bad  ...  \\\n",
-       "0          0.0           1.0           1.0           0.0           1.0  ...   \n",
-       "1          0.0           1.0           1.0           0.0           1.0  ...   \n",
-       "2          0.0           1.0           1.0           0.0           1.0  ...   \n",
-       "3          0.0           1.0           1.0           0.0           1.0  ...   \n",
-       "4          0.0           1.0           1.0           0.0           1.0  ...   \n",
+       "   Var107_is_bad  Var19_is_bad  Var74_is_bad  Var117_is_bad  Var159_is_bad  \\\n",
+       "0            1.0           1.0           0.0            1.0            1.0   \n",
+       "1            1.0           1.0           0.0            1.0            1.0   \n",
+       "2            1.0           1.0           0.0            1.0            1.0   \n",
+       "3            1.0           1.0           0.0            1.0            1.0   \n",
+       "4            1.0           1.0           0.0            1.0            1.0   \n",
        "\n",
-       "   Var227_lev_RAYp  Var227_lev_ZI9m  Var228_logit_code  \\\n",
-       "0              1.0              0.0           0.145563   \n",
-       "1              1.0              0.0           0.150727   \n",
-       "2              0.0              0.0          -0.591072   \n",
-       "3              1.0              0.0           0.150727   \n",
-       "4              0.0              0.0          -0.323715   \n",
+       "   ...  Var191_lev__NA_  Var191_lev_r__I  Var213_logit_code  \\\n",
+       "0  ...              1.0              0.0           0.006417   \n",
+       "1  ...              1.0              0.0           0.008288   \n",
+       "2  ...              1.0              0.0           0.008288   \n",
+       "3  ...              1.0              0.0           0.008288   \n",
+       "4  ...              1.0              0.0           0.008026   \n",
        "\n",
-       "   Var228_prevalence_code  Var228_lev_F2FyR07IdsN7I  Var229_logit_code  \\\n",
-       "0                0.654178                       1.0           0.180634   \n",
-       "1                0.654178                       1.0           0.175825   \n",
-       "2                0.053667                       0.0          -0.296854   \n",
-       "3                0.654178                       1.0          -0.292587   \n",
-       "4                0.018556                       0.0          -0.268261   \n",
+       "   Var213_prevalence_code  Var213_lev__NA_  Var213_lev_KdSa  \\\n",
+       "0                0.977733              1.0              0.0   \n",
+       "1                0.977733              1.0              0.0   \n",
+       "2                0.977733              1.0              0.0   \n",
+       "3                0.977733              1.0              0.0   \n",
+       "4                0.977733              1.0              0.0   \n",
        "\n",
-       "   Var229_prevalence_code  Var229_lev__NA_  Var229_lev_am7c  Var229_lev_mj86  \n",
-       "0                0.568733              1.0              0.0              0.0  \n",
-       "1                0.568733              1.0              0.0              0.0  \n",
-       "2                0.233689              0.0              1.0              0.0  \n",
-       "3                0.196044              0.0              0.0              1.0  \n",
-       "4                0.233689              0.0              1.0              0.0  \n",
+       "   Var201_logit_code  Var201_prevalence_code  Var201_lev__NA_  Var201_lev_smXZ  \n",
+       "0           0.036642                0.744956              1.0              0.0  \n",
+       "1           0.039494                0.744956              1.0              0.0  \n",
+       "2          -0.125971                0.254956              0.0              1.0  \n",
+       "3          -0.125971                0.254956              0.0              1.0  \n",
+       "4           0.036528                0.744956              1.0              0.0  \n",
        "\n",
-       "[5 rows x 233 columns]"
+       "[5 rows x 261 columns]"
       ]
      },
      "execution_count": 13,
@@ -777,7 +789,7 @@
     {
      "data": {
       "text/plain": [
-       "(45000, 233)"
+       "(45000, 261)"
       ]
      },
      "execution_count": 14,
@@ -842,70 +854,70 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>Var1_is_bad</td>\n",
-       "      <td>Var1</td>\n",
+       "      <td>Var11_is_bad</td>\n",
+       "      <td>Var11</td>\n",
        "      <td>missing_indicator</td>\n",
        "      <td>False</td>\n",
        "      <td>True</td>\n",
-       "      <td>0.004328</td>\n",
-       "      <td>0.000037</td>\n",
-       "      <td>0.348710</td>\n",
+       "      <td>0.016325</td>\n",
+       "      <td>0.000576</td>\n",
+       "      <td>2.253129e-04</td>\n",
        "      <td>193.0</td>\n",
        "      <td>0.001036</td>\n",
-       "      <td>False</td>\n",
+       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>Var2_is_bad</td>\n",
-       "      <td>Var2</td>\n",
+       "      <td>Var82_is_bad</td>\n",
+       "      <td>Var82</td>\n",
        "      <td>missing_indicator</td>\n",
        "      <td>False</td>\n",
        "      <td>True</td>\n",
-       "      <td>0.016358</td>\n",
-       "      <td>0.000579</td>\n",
-       "      <td>0.000218</td>\n",
+       "      <td>0.020327</td>\n",
+       "      <td>0.000906</td>\n",
+       "      <td>3.759462e-06</td>\n",
        "      <td>193.0</td>\n",
        "      <td>0.001036</td>\n",
        "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>Var3_is_bad</td>\n",
-       "      <td>Var3</td>\n",
+       "      <td>Var144_is_bad</td>\n",
+       "      <td>Var144</td>\n",
        "      <td>missing_indicator</td>\n",
        "      <td>False</td>\n",
        "      <td>True</td>\n",
-       "      <td>0.016325</td>\n",
-       "      <td>0.000576</td>\n",
-       "      <td>0.000225</td>\n",
+       "      <td>-0.032533</td>\n",
+       "      <td>0.002233</td>\n",
+       "      <td>3.856915e-13</td>\n",
        "      <td>193.0</td>\n",
        "      <td>0.001036</td>\n",
        "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>Var4_is_bad</td>\n",
-       "      <td>Var4</td>\n",
+       "      <td>Var61_is_bad</td>\n",
+       "      <td>Var61</td>\n",
        "      <td>missing_indicator</td>\n",
        "      <td>False</td>\n",
        "      <td>True</td>\n",
-       "      <td>0.020327</td>\n",
-       "      <td>0.000906</td>\n",
-       "      <td>0.000004</td>\n",
+       "      <td>0.014288</td>\n",
+       "      <td>0.000446</td>\n",
+       "      <td>1.169199e-03</td>\n",
        "      <td>193.0</td>\n",
        "      <td>0.001036</td>\n",
-       "      <td>True</td>\n",
+       "      <td>False</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>Var5_is_bad</td>\n",
-       "      <td>Var5</td>\n",
+       "      <td>Var27_is_bad</td>\n",
+       "      <td>Var27</td>\n",
        "      <td>missing_indicator</td>\n",
        "      <td>False</td>\n",
        "      <td>True</td>\n",
        "      <td>0.017267</td>\n",
        "      <td>0.000641</td>\n",
-       "      <td>0.000100</td>\n",
+       "      <td>9.975686e-05</td>\n",
        "      <td>193.0</td>\n",
        "      <td>0.001036</td>\n",
        "      <td>True</td>\n",
@@ -915,19 +927,19 @@
        "</div>"
       ],
       "text/plain": [
-       "      variable orig_variable          treatment  y_aware  has_range  PearsonR  \\\n",
-       "0  Var1_is_bad          Var1  missing_indicator    False       True  0.004328   \n",
-       "1  Var2_is_bad          Var2  missing_indicator    False       True  0.016358   \n",
-       "2  Var3_is_bad          Var3  missing_indicator    False       True  0.016325   \n",
-       "3  Var4_is_bad          Var4  missing_indicator    False       True  0.020327   \n",
-       "4  Var5_is_bad          Var5  missing_indicator    False       True  0.017267   \n",
+       "        variable orig_variable          treatment  y_aware  has_range  \\\n",
+       "0   Var11_is_bad         Var11  missing_indicator    False       True   \n",
+       "1   Var82_is_bad         Var82  missing_indicator    False       True   \n",
+       "2  Var144_is_bad        Var144  missing_indicator    False       True   \n",
+       "3   Var61_is_bad         Var61  missing_indicator    False       True   \n",
+       "4   Var27_is_bad         Var27  missing_indicator    False       True   \n",
        "\n",
-       "         R2  significance  vcount  default_threshold  recommended  \n",
-       "0  0.000037      0.348710   193.0           0.001036        False  \n",
-       "1  0.000579      0.000218   193.0           0.001036         True  \n",
-       "2  0.000576      0.000225   193.0           0.001036         True  \n",
-       "3  0.000906      0.000004   193.0           0.001036         True  \n",
-       "4  0.000641      0.000100   193.0           0.001036         True  "
+       "   PearsonR        R2  significance  vcount  default_threshold  recommended  \n",
+       "0  0.016325  0.000576  2.253129e-04   193.0           0.001036         True  \n",
+       "1  0.020327  0.000906  3.759462e-06   193.0           0.001036         True  \n",
+       "2 -0.032533  0.002233  3.856915e-13   193.0           0.001036         True  \n",
+       "3  0.014288  0.000446  1.169199e-03   193.0           0.001036        False  \n",
+       "4  0.017267  0.000641  9.975686e-05   193.0           0.001036         True  "
       ]
      },
      "execution_count": 15,
@@ -951,7 +963,7 @@
     {
      "data": {
       "text/plain": [
-       "233"
+       "261"
       ]
      },
      "execution_count": 16,
@@ -988,18 +1000,18 @@
     {
      "data": {
       "text/plain": [
-       "Var2_is_bad                            float64\n",
-       "Var3_is_bad                            float64\n",
-       "Var4_is_bad                            float64\n",
-       "Var5_is_bad                            float64\n",
-       "Var6_is_bad                            float64\n",
+       "Var11_is_bad                           float64\n",
+       "Var82_is_bad                           float64\n",
+       "Var144_is_bad                          float64\n",
+       "Var27_is_bad                           float64\n",
+       "Var54_is_bad                           float64\n",
        "                                  ...         \n",
-       "Var229_logit_code                      float64\n",
-       "Var229_prevalence_code                 float64\n",
-       "Var229_lev__NA_           Sparse[float64, 0.0]\n",
-       "Var229_lev_am7c           Sparse[float64, 0.0]\n",
-       "Var229_lev_mj86           Sparse[float64, 0.0]\n",
-       "Length: 233, dtype: object"
+       "Var213_lev_KdSa           Sparse[float64, 0.0]\n",
+       "Var201_logit_code                      float64\n",
+       "Var201_prevalence_code                 float64\n",
+       "Var201_lev__NA_           Sparse[float64, 0.0]\n",
+       "Var201_lev_smXZ           Sparse[float64, 0.0]\n",
+       "Length: 261, dtype: object"
       ]
      },
      "execution_count": 17,
@@ -1019,16 +1031,7 @@
      "is_executing": false
     }
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "DataFrame.dtypes for data must be int, float or bool.\n",
-      "                Did not expect the data types in fields Var191_lev__NA_, Var193_lev_RO12, Var193_lev_2Knk1KF, Var194_lev__NA_, Var194_lev_SEuy, Var195_lev_taul, Var200_lev__NA_, Var201_lev__NA_, Var201_lev_smXZ, Var205_lev_VpdQ, Var206_lev_IYzP, Var206_lev_zm5i, Var206_lev__NA_, Var207_lev_me75fM6ugJ, Var207_lev_7M47J5GA0pTYIFxg5uy, Var210_lev_uKAI, Var211_lev_L84s, Var211_lev_Mtgm, Var212_lev_NhsEn4L, Var212_lev_XfqtO3UdzaXh_, Var213_lev__NA_, Var214_lev__NA_, Var218_lev_cJvF, Var218_lev_UYBR, Var221_lev_oslk, Var221_lev_zCkv, Var225_lev__NA_, Var225_lev_ELof, Var226_lev_FSa2, Var227_lev_RAYp, Var227_lev_ZI9m, Var228_lev_F2FyR07IdsN7I, Var229_lev__NA_, Var229_lev_am7c, Var229_lev_mj86\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# fails due to sparse columns\n",
     "# can also work around this by setting the vtreat parameter 'sparse_indicators' to False\n",
@@ -1042,15 +1045,7 @@
    "cell_type": "code",
    "execution_count": 19,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "no supported conversion for types: (dtype('O'),)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# also fails\n",
     "try:\n",
@@ -1095,7 +1090,7 @@
    },
    "outputs": [],
    "source": [
-    "x_parameters = {\"max_depth\":3, \"objective\":'binary:logistic'}\n",
+    "x_parameters = {\"max_depth\":3, \"objective\":'binary:logistic', \"eval_metric\": 'logloss'}\n",
     "cv = xgboost.cv(x_parameters, fd, num_boost_round=100, verbose_eval=False)"
    ]
   },
@@ -1129,59 +1124,59 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>train-error-mean</th>\n",
-       "      <th>train-error-std</th>\n",
-       "      <th>test-error-mean</th>\n",
-       "      <th>test-error-std</th>\n",
+       "      <th>train-logloss-mean</th>\n",
+       "      <th>train-logloss-std</th>\n",
+       "      <th>test-logloss-mean</th>\n",
+       "      <th>test-logloss-std</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>0.073300</td>\n",
-       "      <td>0.000709</td>\n",
-       "      <td>0.073311</td>\n",
-       "      <td>0.001447</td>\n",
+       "      <td>0.504877</td>\n",
+       "      <td>0.000578</td>\n",
+       "      <td>0.505215</td>\n",
+       "      <td>0.000469</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>0.073322</td>\n",
-       "      <td>0.000741</td>\n",
-       "      <td>0.073333</td>\n",
-       "      <td>0.001415</td>\n",
+       "      <td>0.403280</td>\n",
+       "      <td>0.000908</td>\n",
+       "      <td>0.403873</td>\n",
+       "      <td>0.000906</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>0.073344</td>\n",
-       "      <td>0.000747</td>\n",
-       "      <td>0.073467</td>\n",
-       "      <td>0.001464</td>\n",
+       "      <td>0.342651</td>\n",
+       "      <td>0.001112</td>\n",
+       "      <td>0.343516</td>\n",
+       "      <td>0.001245</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>0.073356</td>\n",
-       "      <td>0.000739</td>\n",
-       "      <td>0.073467</td>\n",
-       "      <td>0.001464</td>\n",
+       "      <td>0.304941</td>\n",
+       "      <td>0.001207</td>\n",
+       "      <td>0.305867</td>\n",
+       "      <td>0.001540</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>0.073356</td>\n",
-       "      <td>0.000739</td>\n",
-       "      <td>0.073444</td>\n",
-       "      <td>0.001450</td>\n",
+       "      <td>0.280804</td>\n",
+       "      <td>0.001457</td>\n",
+       "      <td>0.282171</td>\n",
+       "      <td>0.001653</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "   train-error-mean  train-error-std  test-error-mean  test-error-std\n",
-       "0          0.073300         0.000709         0.073311        0.001447\n",
-       "1          0.073322         0.000741         0.073333        0.001415\n",
-       "2          0.073344         0.000747         0.073467        0.001464\n",
-       "3          0.073356         0.000739         0.073467        0.001464\n",
-       "4          0.073356         0.000739         0.073444        0.001450"
+       "   train-logloss-mean  train-logloss-std  test-logloss-mean  test-logloss-std\n",
+       "0            0.504877           0.000578           0.505215          0.000469\n",
+       "1            0.403280           0.000908           0.403873          0.000906\n",
+       "2            0.342651           0.001112           0.343516          0.001245\n",
+       "3            0.304941           0.001207           0.305867          0.001540\n",
+       "4            0.280804           0.001457           0.282171          0.001653"
       ]
      },
      "execution_count": 23,
@@ -1223,27 +1218,27 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>train-error-mean</th>\n",
-       "      <th>train-error-std</th>\n",
-       "      <th>test-error-mean</th>\n",
-       "      <th>test-error-std</th>\n",
+       "      <th>train-logloss-mean</th>\n",
+       "      <th>train-logloss-std</th>\n",
+       "      <th>test-logloss-mean</th>\n",
+       "      <th>test-logloss-std</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>69</th>\n",
-       "      <td>0.070411</td>\n",
-       "      <td>0.000774</td>\n",
-       "      <td>0.0724</td>\n",
-       "      <td>0.000756</td>\n",
+       "      <th>33</th>\n",
+       "      <td>0.220917</td>\n",
+       "      <td>0.00157</td>\n",
+       "      <td>0.234319</td>\n",
+       "      <td>0.002622</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "    train-error-mean  train-error-std  test-error-mean  test-error-std\n",
-       "69          0.070411         0.000774           0.0724        0.000756"
+       "    train-logloss-mean  train-logloss-std  test-logloss-mean  test-logloss-std\n",
+       "33            0.220917            0.00157           0.234319          0.002622"
       ]
      },
      "execution_count": 24,
@@ -1252,7 +1247,7 @@
     }
    ],
    "source": [
-    "best = cv.loc[cv[\"test-error-mean\"]<= min(cv[\"test-error-mean\"] + 1.0e-9), :]\n",
+    "best = cv.loc[cv[\"test-logloss-mean\"]<= min(cv[\"test-logloss-mean\"] + 1.0e-9), :]\n",
     "best\n",
     "\n"
    ]
@@ -1269,7 +1264,7 @@
     {
      "data": {
       "text/plain": [
-       "69"
+       "33"
       ]
      },
      "execution_count": 25,
@@ -1299,10 +1294,10 @@
        "              gpu_id=None, importance_type='gain', interaction_constraints=None,\n",
        "              learning_rate=None, max_delta_step=None, max_depth=3,\n",
        "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
-       "              n_estimators=69, n_jobs=None, num_parallel_tree=None,\n",
-       "              objective='binary:logistic', random_state=None, reg_alpha=None,\n",
-       "              reg_lambda=None, scale_pos_weight=None, subsample=None,\n",
-       "              tree_method=None, validate_parameters=False, verbosity=None)"
+       "              n_estimators=33, n_jobs=None, num_parallel_tree=None,\n",
+       "              random_state=None, reg_alpha=None, reg_lambda=None,\n",
+       "              scale_pos_weight=None, subsample=None, tree_method=None,\n",
+       "              validate_parameters=None, verbosity=None)"
       ]
      },
      "execution_count": 26,
@@ -1323,7 +1318,23 @@
      "is_executing": false
     }
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/xgboost/sklearn.py:888: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
+      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[10:36:12] WARNING: ../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
+     ]
+    }
+   ],
    "source": [
     "model = fitter.fit(cross_sparse, churn_train)"
    ]
@@ -1366,7 +1377,16 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1379,7 +1399,7 @@
     {
      "data": {
       "text/plain": [
-       "0.804643846890726"
+       "0.7782912822389915"
       ]
      },
      "execution_count": 29,
@@ -1411,7 +1431,16 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1424,7 +1453,7 @@
     {
      "data": {
       "text/plain": [
-       "0.7452235114016007"
+       "0.7320282952343667"
       ]
      },
      "execution_count": 30,
@@ -1473,9 +1502,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_7/lib/python3.7/site-packages/vtreat/vtreat_api.py:265: UserWarning: possibly called transform on same data used to fit\n",
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/vtreat/vtreat_api.py:276: UserWarning: possibly called transform on same data used to fit\n",
       "(this causes over-fit, please use fit_transform() instead)\n",
-      "  \"possibly called transform on same data used to fit\\n\" +\n"
+      "  warnings.warn(\n"
      ]
     }
    ],
@@ -1495,7 +1524,7 @@
     {
      "data": {
       "text/plain": [
-       "229"
+       "257"
       ]
      },
      "execution_count": 32,
@@ -1525,7 +1554,17 @@
      "is_executing": false
     }
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[10:36:51] WARNING: ../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
+      "[10:36:51] WARNING: ../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
+      "[10:36:51] WARNING: ../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
+     ]
+    }
+   ],
    "source": [
     "fd_naive = xgboost.DMatrix(data=naive_sparse, label=churn_train)\n",
     "x_parameters = {\"max_depth\":3, \"objective\":'binary:logistic'}\n",
@@ -1562,35 +1601,27 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>train-error-mean</th>\n",
-       "      <th>train-error-std</th>\n",
-       "      <th>test-error-mean</th>\n",
-       "      <th>test-error-std</th>\n",
+       "      <th>train-logloss-mean</th>\n",
+       "      <th>train-logloss-std</th>\n",
+       "      <th>test-logloss-mean</th>\n",
+       "      <th>test-logloss-std</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>93</th>\n",
-       "      <td>0.048633</td>\n",
-       "      <td>0.000465</td>\n",
-       "      <td>0.058956</td>\n",
-       "      <td>0.001619</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>94</th>\n",
-       "      <td>0.048633</td>\n",
-       "      <td>0.000504</td>\n",
-       "      <td>0.058956</td>\n",
-       "      <td>0.001620</td>\n",
+       "      <th>98</th>\n",
+       "      <td>0.116544</td>\n",
+       "      <td>0.001117</td>\n",
+       "      <td>0.138058</td>\n",
+       "      <td>0.00184</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "    train-error-mean  train-error-std  test-error-mean  test-error-std\n",
-       "93          0.048633         0.000465         0.058956        0.001619\n",
-       "94          0.048633         0.000504         0.058956        0.001620"
+       "    train-logloss-mean  train-logloss-std  test-logloss-mean  test-logloss-std\n",
+       "98            0.116544           0.001117           0.138058           0.00184"
       ]
      },
      "execution_count": 35,
@@ -1599,7 +1630,7 @@
     }
    ],
    "source": [
-    "bestn = cvn.loc[cvn[\"test-error-mean\"] <= min(cvn[\"test-error-mean\"] + 1.0e-9), :]\n",
+    "bestn = cvn.loc[cvn[\"test-logloss-mean\"] <= min(cvn[\"test-logloss-mean\"] + 1.0e-9), :]\n",
     "bestn"
    ]
   },
@@ -1615,7 +1646,7 @@
     {
      "data": {
       "text/plain": [
-       "93"
+       "98"
       ]
      },
      "execution_count": 36,
@@ -1645,10 +1676,10 @@
        "              gpu_id=None, importance_type='gain', interaction_constraints=None,\n",
        "              learning_rate=None, max_delta_step=None, max_depth=3,\n",
        "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
-       "              n_estimators=93, n_jobs=None, num_parallel_tree=None,\n",
-       "              objective='binary:logistic', random_state=None, reg_alpha=None,\n",
-       "              reg_lambda=None, scale_pos_weight=None, subsample=None,\n",
-       "              tree_method=None, validate_parameters=False, verbosity=None)"
+       "              n_estimators=98, n_jobs=None, num_parallel_tree=None,\n",
+       "              random_state=None, reg_alpha=None, reg_lambda=None,\n",
+       "              scale_pos_weight=None, subsample=None, tree_method=None,\n",
+       "              validate_parameters=None, verbosity=None)"
       ]
      },
      "execution_count": 37,
@@ -1669,7 +1700,23 @@
      "is_executing": false
     }
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/xgboost/sklearn.py:888: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
+      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[10:37:00] WARNING: ../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
+     ]
+    }
+   ],
    "source": [
     "modeln = fittern.fit(naive_sparse, churn_train)"
    ]
@@ -1699,7 +1746,16 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1712,7 +1768,7 @@
     {
      "data": {
       "text/plain": [
-       "0.959688845158327"
+       "0.960718624089619"
       ]
      },
      "execution_count": 40,
@@ -1737,7 +1793,16 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1750,7 +1815,7 @@
     {
      "data": {
       "text/plain": [
-       "0.6004655792617093"
+       "0.600843028512341"
       ]
      },
      "execution_count": 41,
@@ -1788,7 +1853,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.9.4"
   },
   "pycharm": {
    "stem_cell": {
diff --git a/Examples/Pipeline/Pipeline_Example.ipynb b/Examples/Pipeline/Pipeline_Example.ipynb
index ae54fe5..cd63e7e 100644
--- a/Examples/Pipeline/Pipeline_Example.ipynb
+++ b/Examples/Pipeline/Pipeline_Example.ipynb
@@ -120,8 +120,115 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "   x_0  x_1  x_2  x_3  x_4  x_5  x_6  x_7  x_8  x_9\n0  c_3  c_1  NaN  r_8  r_3  c_0  c_2  r_3  c_4  c_1\n1  c_3  NaN  r_9  c_3  c_3  c_1  c_2  c_3  c_0  c_3\n2  NaN  c_3  c_4  c_0  c_1  r_0  c_1  r_9  c_3  c_0\n3  c_4  c_2  c_3  r_9  c_0  r_0  r_7  c_1  c_2  r_1\n4  c_2  c_1  r_1  NaN  c_1  c_3  c_4  c_3  c_0  c_0",
-      "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>x_0</th>\n      <th>x_1</th>\n      <th>x_2</th>\n      <th>x_3</th>\n      <th>x_4</th>\n      <th>x_5</th>\n      <th>x_6</th>\n      <th>x_7</th>\n      <th>x_8</th>\n      <th>x_9</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>c_3</td>\n      <td>c_1</td>\n      <td>NaN</td>\n      <td>r_8</td>\n      <td>r_3</td>\n      <td>c_0</td>\n      <td>c_2</td>\n      <td>r_3</td>\n      <td>c_4</td>\n      <td>c_1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>c_3</td>\n      <td>NaN</td>\n      <td>r_9</td>\n      <td>c_3</td>\n      <td>c_3</td>\n      <td>c_1</td>\n      <td>c_2</td>\n      <td>c_3</td>\n      <td>c_0</td>\n      <td>c_3</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>NaN</td>\n      <td>c_3</td>\n      <td>c_4</td>\n      <td>c_0</td>\n      <td>c_1</td>\n      <td>r_0</td>\n      <td>c_1</td>\n      <td>r_9</td>\n      <td>c_3</td>\n      <td>c_0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>c_4</td>\n      <td>c_2</td>\n      <td>c_3</td>\n      <td>r_9</td>\n      <td>c_0</td>\n      <td>r_0</td>\n      <td>r_7</td>\n      <td>c_1</td>\n      <td>c_2</td>\n      <td>r_1</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>c_2</td>\n      <td>c_1</td>\n      <td>r_1</td>\n      <td>NaN</td>\n      <td>c_1</td>\n      <td>c_3</td>\n      <td>c_4</td>\n      <td>c_3</td>\n      <td>c_0</td>\n      <td>c_0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+      "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>x_0</th>\n",
+       "      <th>x_1</th>\n",
+       "      <th>x_2</th>\n",
+       "      <th>x_3</th>\n",
+       "      <th>x_4</th>\n",
+       "      <th>x_5</th>\n",
+       "      <th>x_6</th>\n",
+       "      <th>x_7</th>\n",
+       "      <th>x_8</th>\n",
+       "      <th>x_9</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>r_8</td>\n",
+       "      <td>r_3</td>\n",
+       "      <td>c_0</td>\n",
+       "      <td>c_2</td>\n",
+       "      <td>r_3</td>\n",
+       "      <td>c_4</td>\n",
+       "      <td>c_1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>c_3</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>r_9</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>c_2</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_0</td>\n",
+       "      <td>c_3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_4</td>\n",
+       "      <td>c_0</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>r_0</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>r_9</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>c_4</td>\n",
+       "      <td>c_2</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>r_9</td>\n",
+       "      <td>c_0</td>\n",
+       "      <td>r_0</td>\n",
+       "      <td>r_7</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>c_2</td>\n",
+       "      <td>r_1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>c_2</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>r_1</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>c_1</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_4</td>\n",
+       "      <td>c_3</td>\n",
+       "      <td>c_0</td>\n",
+       "      <td>c_0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   x_0  x_1  x_2  x_3  x_4  x_5  x_6  x_7  x_8  x_9\n",
+       "0  c_3  c_1  NaN  r_8  r_3  c_0  c_2  r_3  c_4  c_1\n",
+       "1  c_3  NaN  r_9  c_3  c_3  c_1  c_2  c_3  c_0  c_3\n",
+       "2  NaN  c_3  c_4  c_0  c_1  r_0  c_1  r_9  c_3  c_0\n",
+       "3  c_4  c_2  c_3  r_9  c_0  r_0  r_7  c_1  c_2  r_1\n",
+       "4  c_2  c_1  r_1  NaN  c_1  c_3  c_4  c_3  c_0  c_0"
+      ]
      },
      "execution_count": 4,
      "metadata": {},
@@ -139,7 +246,14 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0    False\n1     True\n2     True\n3     True\n4    False\ndtype: bool"
+      "text/plain": [
+       "0    False\n",
+       "1     True\n",
+       "2     True\n",
+       "3     True\n",
+       "4    False\n",
+       "dtype: bool"
+      ]
      },
      "execution_count": 5,
      "metadata": {},
@@ -221,7 +335,14 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "GridSearchCV(cv=5,\n             estimator=Pipeline(steps=[('preprocessor',\n                                        __main__.BinomialOutcomeTreatmentP(outcome_target=True, )),\n                                       ('classifier', LogisticRegression())]),\n             param_grid={'classifier__C': [0.1, 1],\n                         'preprocessor__indicator_min_fraction': [0.01, 0.1]})"
+      "text/plain": [
+       "GridSearchCV(cv=5,\n",
+       "             estimator=Pipeline(steps=[('preprocessor',\n",
+       "                                        BinomialOutcomeTreatmentP()),\n",
+       "                                       ('classifier', LogisticRegression())]),\n",
+       "             param_grid={'classifier__C': [0.1, 1],\n",
+       "                         'preprocessor__indicator_min_fraction': [0.01, 0.1]})"
+      ]
      },
      "execution_count": 8,
      "metadata": {},
@@ -239,7 +360,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "{'classifier__C': 0.1, 'preprocessor__indicator_min_fraction': 0.1}"
+      "text/plain": [
+       "{'classifier__C': 0.1, 'preprocessor__indicator_min_fraction': 0.1}"
+      ]
      },
      "execution_count": 9,
      "metadata": {},
@@ -257,7 +380,33 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "{'mean_fit_time': array([0.71360078, 0.28696437, 0.68817482, 0.28684545]),\n 'std_fit_time': array([0.03392186, 0.00624858, 0.01178248, 0.01096785]),\n 'mean_score_time': array([0.07794299, 0.03840861, 0.07571516, 0.0383904 ]),\n 'std_score_time': array([0.00166816, 0.00067049, 0.00099838, 0.00055073]),\n 'param_classifier__C': masked_array(data=[0.1, 0.1, 1, 1],\n              mask=[False, False, False, False],\n        fill_value='?',\n             dtype=object),\n 'param_preprocessor__indicator_min_fraction': masked_array(data=[0.01, 0.1, 0.01, 0.1],\n              mask=[False, False, False, False],\n        fill_value='?',\n             dtype=object),\n 'params': [{'classifier__C': 0.1,\n   'preprocessor__indicator_min_fraction': 0.01},\n  {'classifier__C': 0.1, 'preprocessor__indicator_min_fraction': 0.1},\n  {'classifier__C': 1, 'preprocessor__indicator_min_fraction': 0.01},\n  {'classifier__C': 1, 'preprocessor__indicator_min_fraction': 0.1}],\n 'split0_test_score': array([0.7  , 0.735, 0.66 , 0.72 ]),\n 'split1_test_score': array([0.735, 0.725, 0.725, 0.75 ]),\n 'split2_test_score': array([0.72 , 0.705, 0.675, 0.67 ]),\n 'split3_test_score': array([0.72 , 0.725, 0.71 , 0.715]),\n 'split4_test_score': array([0.68 , 0.685, 0.67 , 0.7  ]),\n 'mean_test_score': array([0.711, 0.715, 0.688, 0.711]),\n 'std_test_score': array([0.01907878, 0.01788854, 0.02501999, 0.02615339]),\n 'rank_test_score': array([2, 1, 4, 3], dtype=int32)}"
+      "text/plain": [
+       "{'mean_fit_time': array([0.73941984, 0.30364056, 0.73936458, 0.31778779]),\n",
+       " 'std_fit_time': array([0.02801503, 0.00888825, 0.04451578, 0.02578008]),\n",
+       " 'mean_score_time': array([0.07852316, 0.03945322, 0.07792797, 0.04070449]),\n",
+       " 'std_score_time': array([0.00106182, 0.00048192, 0.00118028, 0.00239216]),\n",
+       " 'param_classifier__C': masked_array(data=[0.1, 0.1, 1, 1],\n",
+       "              mask=[False, False, False, False],\n",
+       "        fill_value='?',\n",
+       "             dtype=object),\n",
+       " 'param_preprocessor__indicator_min_fraction': masked_array(data=[0.01, 0.1, 0.01, 0.1],\n",
+       "              mask=[False, False, False, False],\n",
+       "        fill_value='?',\n",
+       "             dtype=object),\n",
+       " 'params': [{'classifier__C': 0.1,\n",
+       "   'preprocessor__indicator_min_fraction': 0.01},\n",
+       "  {'classifier__C': 0.1, 'preprocessor__indicator_min_fraction': 0.1},\n",
+       "  {'classifier__C': 1, 'preprocessor__indicator_min_fraction': 0.01},\n",
+       "  {'classifier__C': 1, 'preprocessor__indicator_min_fraction': 0.1}],\n",
+       " 'split0_test_score': array([0.7  , 0.735, 0.66 , 0.72 ]),\n",
+       " 'split1_test_score': array([0.735, 0.725, 0.725, 0.75 ]),\n",
+       " 'split2_test_score': array([0.72 , 0.705, 0.675, 0.67 ]),\n",
+       " 'split3_test_score': array([0.72 , 0.725, 0.71 , 0.715]),\n",
+       " 'split4_test_score': array([0.68 , 0.685, 0.67 , 0.7  ]),\n",
+       " 'mean_test_score': array([0.711, 0.715, 0.688, 0.711]),\n",
+       " 'std_test_score': array([0.01907878, 0.01788854, 0.02501999, 0.02615339]),\n",
+       " 'rank_test_score': array([2, 1, 4, 3], dtype=int32)}"
+      ]
      },
      "execution_count": 10,
      "metadata": {},
@@ -275,7 +424,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0.7150000000000001"
+      "text/plain": [
+       "0.7150000000000001"
+      ]
      },
      "execution_count": 11,
      "metadata": {},
@@ -293,8 +444,91 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "                                                    mean_test_score  \\\nclassifier__C preprocessor__indicator_min_fraction                    \n0.1           0.01                                            0.711   \n              0.10                                            0.715   \n1.0           0.01                                            0.688   \n              0.10                                            0.711   \n\n                                                    rank_test_score  \\\nclassifier__C preprocessor__indicator_min_fraction                    \n0.1           0.01                                                2   \n              0.10                                                1   \n1.0           0.01                                                4   \n              0.10                                                3   \n\n                                                    std_test_score  \nclassifier__C preprocessor__indicator_min_fraction                  \n0.1           0.01                                        0.019079  \n              0.10                                        0.017889  \n1.0           0.01                                        0.025020  \n              0.10                                        0.026153  ",
-      "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></th>\n      <th>mean_test_score</th>\n      <th>rank_test_score</th>\n      <th>std_test_score</th>\n    </tr>\n    <tr>\n      <th>classifier__C</th>\n      <th>preprocessor__indicator_min_fraction</th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th rowspan=\"2\" valign=\"top\">0.1</th>\n      <th>0.01</th>\n      <td>0.711</td>\n      <td>2</td>\n      <td>0.019079</td>\n    </tr>\n    <tr>\n      <th>0.10</th>\n      <td>0.715</td>\n      <td>1</td>\n      <td>0.017889</td>\n    </tr>\n    <tr>\n      <th rowspan=\"2\" valign=\"top\">1.0</th>\n      <th>0.01</th>\n      <td>0.688</td>\n      <td>4</td>\n      <td>0.025020</td>\n    </tr>\n    <tr>\n      <th>0.10</th>\n      <td>0.711</td>\n      <td>3</td>\n      <td>0.026153</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+      "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></th>\n",
+       "      <th>mean_test_score</th>\n",
+       "      <th>rank_test_score</th>\n",
+       "      <th>std_test_score</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>classifier__C</th>\n",
+       "      <th>preprocessor__indicator_min_fraction</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">0.1</th>\n",
+       "      <th>0.01</th>\n",
+       "      <td>0.711</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.019079</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0.10</th>\n",
+       "      <td>0.715</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.017889</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">1.0</th>\n",
+       "      <th>0.01</th>\n",
+       "      <td>0.688</td>\n",
+       "      <td>4</td>\n",
+       "      <td>0.025020</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0.10</th>\n",
+       "      <td>0.711</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0.026153</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                                    mean_test_score  \\\n",
+       "classifier__C preprocessor__indicator_min_fraction                    \n",
+       "0.1           0.01                                            0.711   \n",
+       "              0.10                                            0.715   \n",
+       "1.0           0.01                                            0.688   \n",
+       "              0.10                                            0.711   \n",
+       "\n",
+       "                                                    rank_test_score  \\\n",
+       "classifier__C preprocessor__indicator_min_fraction                    \n",
+       "0.1           0.01                                                2   \n",
+       "              0.10                                                1   \n",
+       "1.0           0.01                                                4   \n",
+       "              0.10                                                3   \n",
+       "\n",
+       "                                                    std_test_score  \n",
+       "classifier__C preprocessor__indicator_min_fraction                  \n",
+       "0.1           0.01                                        0.019079  \n",
+       "              0.10                                        0.017889  \n",
+       "1.0           0.01                                        0.025020  \n",
+       "              0.10                                        0.026153  "
+      ]
      },
      "execution_count": 12,
      "metadata": {},
@@ -334,7 +568,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0.706"
+      "text/plain": [
+       "0.706"
+      ]
      },
      "execution_count": 13,
      "metadata": {},
@@ -363,14 +599,16 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/Users/johnmount/Documents/work/pyvtreat/pkg/vtreat/vtreat_api.py:273: UserWarning: possibly called transform on same data used to fit\n",
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/vtreat/vtreat_api.py:276: UserWarning: possibly called transform on same data used to fit\n",
       "(this causes over-fit, please use fit_transform() instead)\n",
       "  warnings.warn(\n"
      ]
     },
     {
      "data": {
-      "text/plain": "0.786"
+      "text/plain": [
+       "0.786"
+      ]
      },
      "execution_count": 14,
      "metadata": {},
@@ -402,7 +640,20 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0      x_0_lev_c_3\n1      x_0_lev_c_1\n2      x_0_lev_c_4\n3      x_0_lev_c_0\n4      x_0_lev_c_2\n          ...     \n154    x_3_lev_r_6\n155    x_3_lev_r_8\n156    x_3_lev_r_7\n157    x_3_lev_r_2\n158    x_3_lev_r_0\nName: variable, Length: 159, dtype: object"
+      "text/plain": [
+       "0      x_2_lev_c_0\n",
+       "1      x_2_lev_c_4\n",
+       "2      x_2_lev_c_1\n",
+       "3      x_2_lev_c_2\n",
+       "4      x_2_lev_c_3\n",
+       "          ...     \n",
+       "154    x_4_lev_r_2\n",
+       "155    x_4_lev_r_7\n",
+       "156    x_4_lev_r_9\n",
+       "157    x_4_lev_r_0\n",
+       "158    x_4_lev_r_3\n",
+       "Name: variable, Length: 159, dtype: object"
+      ]
      },
      "execution_count": 15,
      "metadata": {},
@@ -427,7 +678,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0.721"
+      "text/plain": [
+       "0.721"
+      ]
      },
      "execution_count": 16,
      "metadata": {},
@@ -464,14 +717,16 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/Users/johnmount/Documents/work/pyvtreat/pkg/vtreat/vtreat_api.py:273: UserWarning: possibly called transform on same data used to fit\n",
+      "/Users/johnmount/opt/anaconda3/envs/ai_academy_3_9/lib/python3.9/site-packages/vtreat/vtreat_api.py:276: UserWarning: possibly called transform on same data used to fit\n",
       "(this causes over-fit, please use fit_transform() instead)\n",
       "  warnings.warn(\n"
      ]
     },
     {
      "data": {
-      "text/plain": "0.755"
+      "text/plain": [
+       "0.755"
+      ]
      },
      "execution_count": 17,
      "metadata": {},
@@ -498,7 +753,64 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0      x_0_lev_c_3\n1      x_0_lev_c_1\n2      x_0_lev_c_4\n3      x_0_lev_c_0\n4      x_0_lev_c_2\n5      x_4_lev_c_2\n6      x_4_lev_c_0\n7      x_4_lev_c_1\n8      x_4_lev_c_3\n9      x_4_lev_c_4\n10    x_4_lev__NA_\n11     x_8_lev_c_2\n12     x_8_lev_c_3\n13     x_8_lev_c_4\n14     x_8_lev_c_1\n15     x_8_lev_c_0\n16     x_5_lev_c_3\n17     x_5_lev_c_4\n18     x_5_lev_c_0\n19     x_5_lev_c_1\n20     x_5_lev_c_2\n21     x_9_lev_c_1\n22     x_9_lev_c_0\n23     x_9_lev_c_2\n24     x_9_lev_c_4\n25     x_9_lev_c_3\n26    x_9_lev__NA_\n27     x_2_lev_c_0\n28     x_2_lev_c_1\n29     x_2_lev_c_4\n30     x_2_lev_c_2\n31     x_2_lev_c_3\n32    x_2_lev__NA_\n33     x_1_lev_c_3\n34     x_1_lev_c_0\n35     x_1_lev_c_1\n36     x_1_lev_c_4\n37     x_1_lev_c_2\n38    x_1_lev__NA_\n39     x_7_lev_c_0\n40     x_7_lev_c_1\n41     x_7_lev_c_4\n42     x_7_lev_c_2\n43     x_7_lev_c_3\n44    x_7_lev__NA_\n45     x_6_lev_c_3\n46     x_6_lev_c_1\n47     x_6_lev_c_4\n48     x_6_lev_c_2\n49     x_6_lev_c_0\n50     x_3_lev_c_1\n51     x_3_lev_c_0\n52     x_3_lev_c_2\n53     x_3_lev_c_4\n54     x_3_lev_c_3\nName: variable, dtype: object"
+      "text/plain": [
+       "0      x_2_lev_c_0\n",
+       "1      x_2_lev_c_4\n",
+       "2      x_2_lev_c_1\n",
+       "3      x_2_lev_c_2\n",
+       "4      x_2_lev_c_3\n",
+       "5     x_2_lev__NA_\n",
+       "6      x_8_lev_c_2\n",
+       "7      x_8_lev_c_3\n",
+       "8      x_8_lev_c_4\n",
+       "9      x_8_lev_c_1\n",
+       "10     x_8_lev_c_0\n",
+       "11     x_9_lev_c_1\n",
+       "12     x_9_lev_c_0\n",
+       "13     x_9_lev_c_2\n",
+       "14     x_9_lev_c_4\n",
+       "15     x_9_lev_c_3\n",
+       "16    x_9_lev__NA_\n",
+       "17     x_7_lev_c_0\n",
+       "18     x_7_lev_c_1\n",
+       "19     x_7_lev_c_4\n",
+       "20     x_7_lev_c_2\n",
+       "21     x_7_lev_c_3\n",
+       "22    x_7_lev__NA_\n",
+       "23     x_0_lev_c_3\n",
+       "24     x_0_lev_c_1\n",
+       "25     x_0_lev_c_4\n",
+       "26     x_0_lev_c_0\n",
+       "27     x_0_lev_c_2\n",
+       "28     x_1_lev_c_3\n",
+       "29     x_1_lev_c_0\n",
+       "30     x_1_lev_c_1\n",
+       "31     x_1_lev_c_4\n",
+       "32     x_1_lev_c_2\n",
+       "33    x_1_lev__NA_\n",
+       "34     x_3_lev_c_1\n",
+       "35     x_3_lev_c_0\n",
+       "36     x_3_lev_c_2\n",
+       "37     x_3_lev_c_4\n",
+       "38     x_3_lev_c_3\n",
+       "39     x_5_lev_c_3\n",
+       "40     x_5_lev_c_4\n",
+       "41     x_5_lev_c_0\n",
+       "42     x_5_lev_c_1\n",
+       "43     x_5_lev_c_2\n",
+       "44     x_6_lev_c_3\n",
+       "45     x_6_lev_c_1\n",
+       "46     x_6_lev_c_4\n",
+       "47     x_6_lev_c_2\n",
+       "48     x_6_lev_c_0\n",
+       "49     x_4_lev_c_2\n",
+       "50     x_4_lev_c_0\n",
+       "51     x_4_lev_c_1\n",
+       "52     x_4_lev_c_3\n",
+       "53     x_4_lev_c_4\n",
+       "54    x_4_lev__NA_\n",
+       "Name: variable, dtype: object"
+      ]
      },
      "execution_count": 18,
      "metadata": {},
@@ -537,7 +849,17 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "Index(['x_0_lev_c_3', 'x_4_logit_code', 'x_4_lev_c_3', 'x_8_logit_code',\n       'x_8_lev_c_2', 'x_8_lev_c_4', 'x_8_lev_c_1', 'x_8_lev_c_0',\n       'x_5_logit_code', 'x_5_lev_c_3', 'x_5_lev_c_4', 'x_5_lev_c_0',\n       'x_5_lev_c_2', 'x_9_logit_code', 'x_9_lev_c_1', 'x_9_lev_c_4',\n       'x_9_lev_c_3', 'x_2_logit_code', 'x_2_lev_c_1', 'x_2_lev_c_2',\n       'x_1_logit_code', 'x_1_lev_c_4', 'x_7_logit_code', 'x_7_lev_c_0',\n       'x_7_lev_c_2', 'x_7_lev_c_3', 'x_6_logit_code', 'x_3_logit_code',\n       'x_3_prevalence_code', 'x_3_lev_c_2', 'x_3_lev_c_3'],\n      dtype='object')"
+      "text/plain": [
+       "Index(['x_2_logit_code', 'x_2_lev_c_1', 'x_2_lev_c_2', 'x_8_logit_code',\n",
+       "       'x_8_lev_c_2', 'x_8_lev_c_4', 'x_8_lev_c_1', 'x_8_lev_c_0',\n",
+       "       'x_9_logit_code', 'x_9_lev_c_1', 'x_9_lev_c_4', 'x_9_lev_c_3',\n",
+       "       'x_7_logit_code', 'x_7_lev_c_0', 'x_7_lev_c_2', 'x_7_lev_c_3',\n",
+       "       'x_0_lev_c_3', 'x_1_logit_code', 'x_1_lev_c_4', 'x_3_logit_code',\n",
+       "       'x_3_prevalence_code', 'x_3_lev_c_2', 'x_3_lev_c_3', 'x_5_logit_code',\n",
+       "       'x_5_lev_c_3', 'x_5_lev_c_4', 'x_5_lev_c_0', 'x_5_lev_c_2',\n",
+       "       'x_6_logit_code', 'x_4_logit_code', 'x_4_lev_c_3'],\n",
+       "      dtype='object')"
+      ]
      },
      "execution_count": 19,
      "metadata": {},
@@ -565,7 +887,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "{'classifier__C': 0.1}"
+      "text/plain": [
+       "{'classifier__C': 0.1}"
+      ]
      },
      "execution_count": 20,
      "metadata": {},
@@ -595,7 +919,9 @@
    "outputs": [
     {
      "data": {
-      "text/plain": "0.713"
+      "text/plain": [
+       "0.713"
+      ]
      },
      "execution_count": 21,
      "metadata": {},
@@ -619,7 +945,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": []
@@ -641,7 +967,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.9.4"
   },
   "pycharm": {
    "stem_cell": {
@@ -655,4 +981,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
\ No newline at end of file
+}