EDA: Automated Quick Model Fit
==============================

The purpose of this feature is to provide a quick and easy way to obtain
a preliminary understanding of the relationships between the target
variable and the independent variables in a dataset.

This functionality automatically splits the training data, fits a simple
regression or classification model to the data and generates insights:
model performance metrics, feature importance and prediction result
insights.

To inspect the prediction quality, a confusion matrix is displayed for
classification problems and scatter plot for regression problems. Both
representation allow the user to see the difference between actual and
predicted values.

The insights highlight two subsets of the model predictions:

-  predictions with the largest classification error. Rows listed in
   this section are candidates for inspecting why the model made the
   mistakes
-  predictions with the least distance from the other class. Rows in
   this category are most ‘undecided’. They are useful as an examples of
   data which is close to a decision boundary between the classes. The
   model would benefit from having more data for similar cases.

Classification Example
----------------------

We will start with getting titanic dataset and performing a quick
one-line overview to get the information.

.. code:: python

    import pandas as pd
    import autogluon.eda.auto as auto
    
    df_train = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/titanic/train.csv')
    df_test = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/titanic/test.csv')
    target_col = 'Survived'
    
    auto.quick_fit(df_train, target_col, show_feature_importance_barplots=True)


.. parsed-literal::
    :class: output

    No path specified. Models will be saved in: "AutogluonModels/ag-20230204_010022/"



Model Prediction for Survived
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



.. figure:: output_eda-auto-quick-fit_abf798_1_2.png



Model Leaderboard
~~~~~~~~~~~~~~~~~



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>model</th>
          <th>score_test</th>
          <th>score_val</th>
          <th>pred_time_test</th>
          <th>pred_time_val</th>
          <th>fit_time</th>
          <th>pred_time_test_marginal</th>
          <th>pred_time_val_marginal</th>
          <th>fit_time_marginal</th>
          <th>stack_level</th>
          <th>can_infer</th>
          <th>fit_order</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>LightGBMXT</td>
          <td>0.809701</td>
          <td>0.856</td>
          <td>0.004289</td>
          <td>0.003677</td>
          <td>0.769681</td>
          <td>0.004289</td>
          <td>0.003677</td>
          <td>0.769681</td>
          <td>1</td>
          <td>True</td>
          <td>1</td>
        </tr>
      </tbody>
    </table>
    </div>



Feature Importance for Trained Model
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>importance</th>
          <th>stddev</th>
          <th>p_value</th>
          <th>n</th>
          <th>p99_high</th>
          <th>p99_low</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>Sex</th>
          <td>0.112687</td>
          <td>0.013033</td>
          <td>0.000021</td>
          <td>5</td>
          <td>0.139522</td>
          <td>0.085851</td>
        </tr>
        <tr>
          <th>Name</th>
          <td>0.055970</td>
          <td>0.009140</td>
          <td>0.000082</td>
          <td>5</td>
          <td>0.074789</td>
          <td>0.037151</td>
        </tr>
        <tr>
          <th>SibSp</th>
          <td>0.026119</td>
          <td>0.010554</td>
          <td>0.002605</td>
          <td>5</td>
          <td>0.047850</td>
          <td>0.004389</td>
        </tr>
        <tr>
          <th>Fare</th>
          <td>0.012687</td>
          <td>0.009730</td>
          <td>0.021720</td>
          <td>5</td>
          <td>0.032721</td>
          <td>-0.007348</td>
        </tr>
        <tr>
          <th>Embarked</th>
          <td>0.011194</td>
          <td>0.006981</td>
          <td>0.011525</td>
          <td>5</td>
          <td>0.025567</td>
          <td>-0.003179</td>
        </tr>
        <tr>
          <th>Age</th>
          <td>0.010448</td>
          <td>0.003122</td>
          <td>0.000853</td>
          <td>5</td>
          <td>0.016876</td>
          <td>0.004020</td>
        </tr>
        <tr>
          <th>PassengerId</th>
          <td>0.008955</td>
          <td>0.005659</td>
          <td>0.012022</td>
          <td>5</td>
          <td>0.020607</td>
          <td>-0.002696</td>
        </tr>
        <tr>
          <th>Cabin</th>
          <td>0.002985</td>
          <td>0.006675</td>
          <td>0.186950</td>
          <td>5</td>
          <td>0.016729</td>
          <td>-0.010758</td>
        </tr>
        <tr>
          <th>Pclass</th>
          <td>0.002239</td>
          <td>0.005659</td>
          <td>0.213159</td>
          <td>5</td>
          <td>0.013890</td>
          <td>-0.009413</td>
        </tr>
        <tr>
          <th>Parch</th>
          <td>0.001493</td>
          <td>0.002044</td>
          <td>0.088904</td>
          <td>5</td>
          <td>0.005701</td>
          <td>-0.002716</td>
        </tr>
        <tr>
          <th>Ticket</th>
          <td>0.000000</td>
          <td>0.000000</td>
          <td>0.500000</td>
          <td>5</td>
          <td>0.000000</td>
          <td>0.000000</td>
        </tr>
      </tbody>
    </table>
    </div>



.. figure:: output_eda-auto-quick-fit_abf798_1_7.png



Rows with the highest prediction error
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Rows in this category worth inspecting for the causes of the error



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>PassengerId</th>
          <th>Pclass</th>
          <th>Name</th>
          <th>Sex</th>
          <th>Age</th>
          <th>SibSp</th>
          <th>Parch</th>
          <th>Ticket</th>
          <th>Fare</th>
          <th>Cabin</th>
          <th>Embarked</th>
          <th>Survived</th>
          <th>0</th>
          <th>1</th>
          <th>error</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>498</th>
          <td>499</td>
          <td>1</td>
          <td>Allison, Mrs. Hudson J C (Bessie Waldo Daniels)</td>
          <td>female</td>
          <td>25.0</td>
          <td>1</td>
          <td>2</td>
          <td>113781</td>
          <td>151.5500</td>
          <td>C22 C26</td>
          <td>S</td>
          <td>0</td>
          <td>0.046788</td>
          <td>0.953212</td>
          <td>0.953212</td>
        </tr>
        <tr>
          <th>267</th>
          <td>268</td>
          <td>3</td>
          <td>Persson, Mr. Ernst Ulrik</td>
          <td>male</td>
          <td>25.0</td>
          <td>1</td>
          <td>0</td>
          <td>347083</td>
          <td>7.7750</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.932024</td>
          <td>0.067976</td>
          <td>0.932024</td>
        </tr>
        <tr>
          <th>569</th>
          <td>570</td>
          <td>3</td>
          <td>Jonsson, Mr. Carl</td>
          <td>male</td>
          <td>32.0</td>
          <td>0</td>
          <td>0</td>
          <td>350417</td>
          <td>7.8542</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.922265</td>
          <td>0.077735</td>
          <td>0.922265</td>
        </tr>
        <tr>
          <th>283</th>
          <td>284</td>
          <td>3</td>
          <td>Dorking, Mr. Edward Arthur</td>
          <td>male</td>
          <td>19.0</td>
          <td>0</td>
          <td>0</td>
          <td>A/5. 10482</td>
          <td>8.0500</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.921180</td>
          <td>0.078820</td>
          <td>0.921180</td>
        </tr>
        <tr>
          <th>821</th>
          <td>822</td>
          <td>3</td>
          <td>Lulic, Mr. Nikola</td>
          <td>male</td>
          <td>27.0</td>
          <td>0</td>
          <td>0</td>
          <td>315098</td>
          <td>8.6625</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.919709</td>
          <td>0.080291</td>
          <td>0.919709</td>
        </tr>
        <tr>
          <th>301</th>
          <td>302</td>
          <td>3</td>
          <td>McCoy, Mr. Bernard</td>
          <td>male</td>
          <td>NaN</td>
          <td>2</td>
          <td>0</td>
          <td>367226</td>
          <td>23.2500</td>
          <td>NaN</td>
          <td>Q</td>
          <td>1</td>
          <td>0.918546</td>
          <td>0.081454</td>
          <td>0.918546</td>
        </tr>
        <tr>
          <th>288</th>
          <td>289</td>
          <td>2</td>
          <td>Hosono, Mr. Masabumi</td>
          <td>male</td>
          <td>42.0</td>
          <td>0</td>
          <td>0</td>
          <td>237798</td>
          <td>13.0000</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.907043</td>
          <td>0.092957</td>
          <td>0.907043</td>
        </tr>
        <tr>
          <th>36</th>
          <td>37</td>
          <td>3</td>
          <td>Mamee, Mr. Hanna</td>
          <td>male</td>
          <td>NaN</td>
          <td>0</td>
          <td>0</td>
          <td>2677</td>
          <td>7.2292</td>
          <td>NaN</td>
          <td>C</td>
          <td>1</td>
          <td>0.906803</td>
          <td>0.093197</td>
          <td>0.906803</td>
        </tr>
        <tr>
          <th>127</th>
          <td>128</td>
          <td>3</td>
          <td>Madsen, Mr. Fridtjof Arne</td>
          <td>male</td>
          <td>24.0</td>
          <td>0</td>
          <td>0</td>
          <td>C 17369</td>
          <td>7.1417</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.906605</td>
          <td>0.093395</td>
          <td>0.906605</td>
        </tr>
        <tr>
          <th>391</th>
          <td>392</td>
          <td>3</td>
          <td>Jansson, Mr. Carl Olof</td>
          <td>male</td>
          <td>21.0</td>
          <td>0</td>
          <td>0</td>
          <td>350034</td>
          <td>7.7958</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.905367</td>
          <td>0.094633</td>
          <td>0.905367</td>
        </tr>
      </tbody>
    </table>
    </div>



Rows with the least distance vs other class
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Rows in this category are the closest to the decision boundary vs the
other class and are good candidates for additional labeling



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>PassengerId</th>
          <th>Pclass</th>
          <th>Name</th>
          <th>Sex</th>
          <th>Age</th>
          <th>SibSp</th>
          <th>Parch</th>
          <th>Ticket</th>
          <th>Fare</th>
          <th>Cabin</th>
          <th>Embarked</th>
          <th>Survived</th>
          <th>0</th>
          <th>1</th>
          <th>score_diff</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>182</th>
          <td>183</td>
          <td>3</td>
          <td>Asplund, Master. Clarence Gustaf Hugo</td>
          <td>male</td>
          <td>9.0</td>
          <td>4</td>
          <td>2</td>
          <td>347077</td>
          <td>31.3875</td>
          <td>NaN</td>
          <td>S</td>
          <td>0</td>
          <td>0.503872</td>
          <td>0.496128</td>
          <td>0.007743</td>
        </tr>
        <tr>
          <th>475</th>
          <td>476</td>
          <td>1</td>
          <td>Clifford, Mr. George Quincy</td>
          <td>male</td>
          <td>NaN</td>
          <td>0</td>
          <td>0</td>
          <td>110465</td>
          <td>52.0000</td>
          <td>A14</td>
          <td>S</td>
          <td>0</td>
          <td>0.509178</td>
          <td>0.490822</td>
          <td>0.018356</td>
        </tr>
        <tr>
          <th>347</th>
          <td>348</td>
          <td>3</td>
          <td>Davison, Mrs. Thomas Henry (Mary E Finck)</td>
          <td>female</td>
          <td>NaN</td>
          <td>1</td>
          <td>0</td>
          <td>386525</td>
          <td>16.1000</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.510786</td>
          <td>0.489214</td>
          <td>0.021572</td>
        </tr>
        <tr>
          <th>192</th>
          <td>193</td>
          <td>3</td>
          <td>Andersen-Jensen, Miss. Carla Christine Nielsine</td>
          <td>female</td>
          <td>19.0</td>
          <td>1</td>
          <td>0</td>
          <td>350046</td>
          <td>7.8542</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.512167</td>
          <td>0.487833</td>
          <td>0.024334</td>
        </tr>
        <tr>
          <th>330</th>
          <td>331</td>
          <td>3</td>
          <td>McCoy, Miss. Agnes</td>
          <td>female</td>
          <td>NaN</td>
          <td>2</td>
          <td>0</td>
          <td>367226</td>
          <td>23.2500</td>
          <td>NaN</td>
          <td>Q</td>
          <td>1</td>
          <td>0.478502</td>
          <td>0.521498</td>
          <td>0.042996</td>
        </tr>
        <tr>
          <th>572</th>
          <td>573</td>
          <td>1</td>
          <td>Flynn, Mr. John Irwin ("Irving")</td>
          <td>male</td>
          <td>36.0</td>
          <td>0</td>
          <td>0</td>
          <td>PC 17474</td>
          <td>26.3875</td>
          <td>E25</td>
          <td>S</td>
          <td>1</td>
          <td>0.478234</td>
          <td>0.521766</td>
          <td>0.043532</td>
        </tr>
        <tr>
          <th>792</th>
          <td>793</td>
          <td>3</td>
          <td>Sage, Miss. Stella Anna</td>
          <td>female</td>
          <td>NaN</td>
          <td>8</td>
          <td>2</td>
          <td>CA. 2343</td>
          <td>69.5500</td>
          <td>NaN</td>
          <td>S</td>
          <td>0</td>
          <td>0.525041</td>
          <td>0.474959</td>
          <td>0.050082</td>
        </tr>
        <tr>
          <th>172</th>
          <td>173</td>
          <td>3</td>
          <td>Johnson, Miss. Eleanor Ileen</td>
          <td>female</td>
          <td>1.0</td>
          <td>1</td>
          <td>1</td>
          <td>347742</td>
          <td>11.1333</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.526793</td>
          <td>0.473207</td>
          <td>0.053585</td>
        </tr>
        <tr>
          <th>328</th>
          <td>329</td>
          <td>3</td>
          <td>Goldsmith, Mrs. Frank John (Emily Alice Brown)</td>
          <td>female</td>
          <td>31.0</td>
          <td>1</td>
          <td>1</td>
          <td>363291</td>
          <td>20.5250</td>
          <td>NaN</td>
          <td>S</td>
          <td>1</td>
          <td>0.531574</td>
          <td>0.468426</td>
          <td>0.063149</td>
        </tr>
        <tr>
          <th>593</th>
          <td>594</td>
          <td>3</td>
          <td>Bourke, Miss. Mary</td>
          <td>female</td>
          <td>NaN</td>
          <td>0</td>
          <td>2</td>
          <td>364848</td>
          <td>7.7500</td>
          <td>NaN</td>
          <td>Q</td>
          <td>0</td>
          <td>0.463840</td>
          <td>0.536160</td>
          <td>0.072319</td>
        </tr>
      </tbody>
    </table>
    </div>


Regression Example
------------------

In the previous section we tried a classification example. Let’s try a
regression. It has a few differences. We are also going to store the
fitted model by specifying ``return_state`` and ``save_model_to_state``
parameters. This will allow us to use the model to predict test values
later.

It is a large dataset, so we’ll keep only a few columns for this
tutorial.

.. code:: python

    df_train = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/AmesHousingPriceRegression/train_data.csv')
    df_test = pd.read_csv('https://autogluon.s3.amazonaws.com/datasets/AmesHousingPriceRegression/test_data.csv')
    target_col = 'SalePrice'
    
    keep_cols = [
      'Overall.Qual', 'Gr.Liv.Area', 'Neighborhood', 'Total.Bsmt.SF', 'BsmtFin.SF.1',
      'X1st.Flr.SF', 'Bsmt.Qual', 'Garage.Cars', 'Half.Bath', 'Year.Remod.Add', target_col
    ]
    
    df_train = df_train[[c for c in df_train.columns if c in keep_cols]][:500]
    df_test = df_test[[c for c in df_test.columns if c in keep_cols]][:500]
    
    
    state = auto.quick_fit(df_train, target_col, return_state=True, save_model_to_state=True)


.. parsed-literal::
    :class: output

    No path specified. Models will be saved in: "AutogluonModels/ag-20230204_010025/"



Model Prediction for SalePrice
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



.. figure:: output_eda-auto-quick-fit_abf798_3_2.png



Model Leaderboard
~~~~~~~~~~~~~~~~~



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>model</th>
          <th>score_test</th>
          <th>score_val</th>
          <th>pred_time_test</th>
          <th>pred_time_val</th>
          <th>fit_time</th>
          <th>pred_time_test_marginal</th>
          <th>pred_time_val_marginal</th>
          <th>fit_time_marginal</th>
          <th>stack_level</th>
          <th>can_infer</th>
          <th>fit_order</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>LightGBMXT</td>
          <td>-29100.820216</td>
          <td>-31075.785774</td>
          <td>0.003136</td>
          <td>0.003143</td>
          <td>0.722433</td>
          <td>0.003136</td>
          <td>0.003143</td>
          <td>0.722433</td>
          <td>1</td>
          <td>True</td>
          <td>1</td>
        </tr>
      </tbody>
    </table>
    </div>



Feature Importance for Trained Model
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>importance</th>
          <th>stddev</th>
          <th>p_value</th>
          <th>n</th>
          <th>p99_high</th>
          <th>p99_low</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>Overall.Qual</th>
          <td>16126.273271</td>
          <td>1376.470881</td>
          <td>0.000006</td>
          <td>5</td>
          <td>18960.445840</td>
          <td>13292.100702</td>
        </tr>
        <tr>
          <th>Gr.Liv.Area</th>
          <td>8862.693281</td>
          <td>480.183424</td>
          <td>0.000001</td>
          <td>5</td>
          <td>9851.397587</td>
          <td>7873.988974</td>
        </tr>
        <tr>
          <th>Total.Bsmt.SF</th>
          <td>5299.844900</td>
          <td>870.222500</td>
          <td>0.000084</td>
          <td>5</td>
          <td>7091.645055</td>
          <td>3508.044746</td>
        </tr>
        <tr>
          <th>Garage.Cars</th>
          <td>4472.147453</td>
          <td>660.340484</td>
          <td>0.000055</td>
          <td>5</td>
          <td>5831.797636</td>
          <td>3112.497270</td>
        </tr>
        <tr>
          <th>X1st.Flr.SF</th>
          <td>3804.848692</td>
          <td>692.065035</td>
          <td>0.000126</td>
          <td>5</td>
          <td>5229.820166</td>
          <td>2379.877219</td>
        </tr>
        <tr>
          <th>BsmtFin.SF.1</th>
          <td>3725.145846</td>
          <td>369.988099</td>
          <td>0.000012</td>
          <td>5</td>
          <td>4486.956454</td>
          <td>2963.335237</td>
        </tr>
        <tr>
          <th>Year.Remod.Add</th>
          <td>3562.868687</td>
          <td>1081.770172</td>
          <td>0.000906</td>
          <td>5</td>
          <td>5790.248423</td>
          <td>1335.488951</td>
        </tr>
        <tr>
          <th>Half.Bath</th>
          <td>3020.213571</td>
          <td>1041.365031</td>
          <td>0.001457</td>
          <td>5</td>
          <td>5164.398563</td>
          <td>876.028580</td>
        </tr>
        <tr>
          <th>Neighborhood</th>
          <td>624.378438</td>
          <td>297.685200</td>
          <td>0.004689</td>
          <td>5</td>
          <td>1237.316379</td>
          <td>11.440497</td>
        </tr>
        <tr>
          <th>Bsmt.Qual</th>
          <td>0.000000</td>
          <td>0.000000</td>
          <td>0.500000</td>
          <td>5</td>
          <td>0.000000</td>
          <td>0.000000</td>
        </tr>
      </tbody>
    </table>
    </div>



Rows with the highest prediction error
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Rows in this category worth inspecting for the causes of the error



.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Neighborhood</th>
          <th>Overall.Qual</th>
          <th>Year.Remod.Add</th>
          <th>Bsmt.Qual</th>
          <th>BsmtFin.SF.1</th>
          <th>Total.Bsmt.SF</th>
          <th>X1st.Flr.SF</th>
          <th>Gr.Liv.Area</th>
          <th>Half.Bath</th>
          <th>Garage.Cars</th>
          <th>SalePrice</th>
          <th>SalePrice_pred</th>
          <th>error</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>134</th>
          <td>Edwards</td>
          <td>6</td>
          <td>1966</td>
          <td>Gd</td>
          <td>0.0</td>
          <td>697.0</td>
          <td>1575</td>
          <td>2201</td>
          <td>0</td>
          <td>2.0</td>
          <td>274970</td>
          <td>150482.062500</td>
          <td>124487.937500</td>
        </tr>
        <tr>
          <th>90</th>
          <td>Timber</td>
          <td>10</td>
          <td>2007</td>
          <td>Ex</td>
          <td>0.0</td>
          <td>1824.0</td>
          <td>1824</td>
          <td>1824</td>
          <td>0</td>
          <td>3.0</td>
          <td>392000</td>
          <td>277549.312500</td>
          <td>114450.687500</td>
        </tr>
        <tr>
          <th>468</th>
          <td>NridgHt</td>
          <td>9</td>
          <td>2003</td>
          <td>Ex</td>
          <td>1972.0</td>
          <td>2452.0</td>
          <td>2452</td>
          <td>2452</td>
          <td>0</td>
          <td>3.0</td>
          <td>445000</td>
          <td>344158.687500</td>
          <td>100841.312500</td>
        </tr>
        <tr>
          <th>45</th>
          <td>NridgHt</td>
          <td>9</td>
          <td>2006</td>
          <td>Ex</td>
          <td>0.0</td>
          <td>1704.0</td>
          <td>1722</td>
          <td>2758</td>
          <td>1</td>
          <td>3.0</td>
          <td>418000</td>
          <td>327791.531250</td>
          <td>90208.468750</td>
        </tr>
        <tr>
          <th>118</th>
          <td>Somerst</td>
          <td>7</td>
          <td>2006</td>
          <td>Gd</td>
          <td>788.0</td>
          <td>960.0</td>
          <td>960</td>
          <td>2318</td>
          <td>1</td>
          <td>2.0</td>
          <td>294323</td>
          <td>218767.953125</td>
          <td>75555.046875</td>
        </tr>
        <tr>
          <th>318</th>
          <td>Crawfor</td>
          <td>7</td>
          <td>2002</td>
          <td>Gd</td>
          <td>1406.0</td>
          <td>1902.0</td>
          <td>1902</td>
          <td>1902</td>
          <td>0</td>
          <td>2.0</td>
          <td>335000</td>
          <td>265314.125000</td>
          <td>69685.875000</td>
        </tr>
        <tr>
          <th>26</th>
          <td>Mitchel</td>
          <td>5</td>
          <td>2006</td>
          <td>NaN</td>
          <td>0.0</td>
          <td>0.0</td>
          <td>1771</td>
          <td>1771</td>
          <td>0</td>
          <td>2.0</td>
          <td>115000</td>
          <td>179189.171875</td>
          <td>64189.171875</td>
        </tr>
        <tr>
          <th>233</th>
          <td>NoRidge</td>
          <td>8</td>
          <td>2000</td>
          <td>Gd</td>
          <td>655.0</td>
          <td>1145.0</td>
          <td>1145</td>
          <td>2198</td>
          <td>1</td>
          <td>3.0</td>
          <td>250000</td>
          <td>311554.031250</td>
          <td>61554.031250</td>
        </tr>
        <tr>
          <th>322</th>
          <td>NoRidge</td>
          <td>8</td>
          <td>1993</td>
          <td>Gd</td>
          <td>1129.0</td>
          <td>1390.0</td>
          <td>1402</td>
          <td>2225</td>
          <td>1</td>
          <td>3.0</td>
          <td>285000</td>
          <td>341849.781250</td>
          <td>56849.781250</td>
        </tr>
        <tr>
          <th>340</th>
          <td>ClearCr</td>
          <td>7</td>
          <td>2005</td>
          <td>Gd</td>
          <td>226.0</td>
          <td>1385.0</td>
          <td>1363</td>
          <td>1363</td>
          <td>0</td>
          <td>2.0</td>
          <td>241500</td>
          <td>185450.109375</td>
          <td>56049.890625</td>
        </tr>
      </tbody>
    </table>
    </div>


Using a fitted model
--------------------

Now let’s get the ``model`` from ``state``, perform the prediction on
``df_test`` and quickly visualize the results using
``auto.analyze_interaction()`` tool:

.. code:: python

    model = state.model
    y_pred = model.predict(df_test)
    auto.analyze_interaction(
        train_data=pd.DataFrame({'SalePrice_Pred': y_pred}), 
        x='SalePrice_Pred', 
        fit_distributions=['johnsonsu', 'norm', 'exponnorm']
    )



.. figure:: output_eda-auto-quick-fit_abf798_5_0.png