# Training ML Models

In this guide, we are going to use TQL to train a simple model that predicts the probability we will
win an advertising auction, given the ad-size.  We will then publish that model back into TQL and
utilize it in a query that displays model predictions next to ad-size.

## Inspect The bid TimeSeries
Use a simple `SELECT bid.*` statement to inspect the fields and the values of the bid time series:


```python
from zeenk.tql import *

select('bid.*').from_events('lethe4').limit(10)
```

    ------------------------------------------------------------------
               Version: 20.1.17-SNAPSHOT
     Version Timestamp: 2021-08-27 20:56:17
           Version Age: 3 days, 13 hours, 38 minutes, 14 seconds
       Filesystem Root: /Users/zkozick/.tql/files
     Working Directory: /Users/zkozick/.tql
    Configuration File: /Users/zkozick/.tql/tql-conf.yml
           Api Gateway: http://localhost:9000
        Service Status: Icarus: ONLINE, Daedalus: ONLINE
        Service Uptime: 18 hours, 6 minutes, 13 seconds
    ------------------------------------------------------------------





<h4><em>Query results:</em></h4>partition "_default"
<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>bid_ad_size</th>
      <th>bid_bid</th>
      <th>bid_event_time</th>
      <th>bid_ghosted</th>
      <th>bid_request_id</th>
      <th>bid_user_id</th>
      <th>bid_won</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>1</th>
      <td>big</td>
      <td>1.0</td>
      <td>2021-01-02 00:02:36</td>
      <td>false</td>
      <td>r13260</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>2</th>
      <td>big</td>
      <td>1.0</td>
      <td>2021-01-02 00:16:41</td>
      <td>false</td>
      <td>r13376</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>3</th>
      <td>small</td>
      <td>1.0</td>
      <td>2021-01-02 00:30:55</td>
      <td>false</td>
      <td>r13509</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>4</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>5</th>
      <td>small</td>
      <td>1.0</td>
      <td>2021-01-02 01:04:01</td>
      <td>false</td>
      <td>r13737</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>6</th>
      <td>big</td>
      <td>1.0</td>
      <td>2021-01-02 01:37:26</td>
      <td>false</td>
      <td>r13985</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>7</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>8</th>
      <td>big</td>
      <td>1.0</td>
      <td>2021-01-02 02:18:30</td>
      <td>false</td>
      <td>r14310</td>
      <td>u252</td>
      <td>true</td>
    </tr>
    <tr>
      <th>9</th>
      <td>big</td>
      <td>1.0</td>
      <td>2021-01-02 03:01:48</td>
      <td>false</td>
      <td>r14607</td>
      <td>u252</td>
      <td>true</td>
    </tr>
  </tbody>
</table>
</div><i>query produced 10 rows x 7 columns in 0.51 seconds</i><br><br>



## Create a Training ResultSet

Let's create a ResultSet using the fields of `bid` event.  For this TQL query, we cast the columns to machine
learning types using the function `label()`, `weight()`, and `categorical()`.  Note there is also `numerical()`,
however, we are not using it in this model as the `bid` event object has no fields that could be used
as numerical features.

An important aspect of creating machine learning ResultSets is the `partition_by()` statement.  `partition_by()`
takes a TQL expression which emits an enumerated string for each row.  The output dataframe is split into partitions
based on the value of the partition key expression for that row.  This allows you to produce reproducible
train/test splits over multiple runs, if your expression is based on an attribute of the event, as
follows.


```python
result_set = select(
    label('if(bid.won, 1, 0)'),
    weight(1.0),
    categorical('lower(bid.ad_size)', 'ad_size')
)\
.from_events('lethe4')\
.where('type == "bid"')\
.partition_by('IF(HASH(bid.user_id) % 10 < 2, "test", "train")')\
.options(expand_numerical_features=True, fill_na=True)\
.submit()

result_set
```




<h4><em>Query results:</em></h4>partition "test"
<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>_label</th>
      <th>_weight</th>
      <th>ad_size</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>1</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>2</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>3</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>4</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>...</th>
      <td>...</td>
      <td>...</td>
      <td>...</td>
    </tr>
    <tr>
      <th>33554</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>33555</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>33556</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>33557</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>33558</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
  </tbody>
</table>
<p>33559 rows × 3 columns</p>
</div>partition "train"
<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>_label</th>
      <th>_weight</th>
      <th>ad_size</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>1</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>2</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>3</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>4</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>...</th>
      <td>...</td>
      <td>...</td>
      <td>...</td>
    </tr>
    <tr>
      <th>138413</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>138414</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>138415</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>big</td>
    </tr>
    <tr>
      <th>138416</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
    <tr>
      <th>138417</th>
      <td>1.0</td>
      <td>1.0</td>
      <td>small</td>
    </tr>
  </tbody>
</table>
<p>138418 rows × 3 columns</p>
</div><i>query produced 171977 rows x 3 columns in 8.48 seconds</i><br><br>



## Estimating a Model

We now pass the ResultSet directly into a TQL Estimator in order to produce a trained model. TQL
provides convenience wrapper for H2O, but drop-in support for other training packages can be made
available [upon request](mailto:zkozick@nanigans.com?subject=Request%20For%20Additional%20TQL%20Training%20Engine%20Support).

When training, a `TuningConf` and/or `BootstrapConf` may be provided.  `TuningConf` enables hyper parameter
tuning during the training session.  `BootstrapConf` trains an ensemble of models with perturbed
weights, enabling the estimation of standard errors via a
[Bayesian Bootstrap](https://gdmarmerola.github.io/the-bayesian-bootstrap/) technique.

The estimator emits a PublishedModel instance, which we will then inspect.


```python
from zeenk.tql.modeling import H2OEstimator, TuningConf, BootstrapConf

estimator = H2OEstimator('linear', result_set, 'glm')
estimator.get_tag()
model = estimator.train(
    tuning_conf=TuningConf(init_pts=10, iterations=10),
    bootstrap_conf=BootstrapConf(bootstraps=10)
)
```

    Checking whether there is an H2O instance running at http://localhost:54321 ..... not found.
    Attempting to start a local H2O server...
      Java Version: openjdk version "1.8.0_275"; OpenJDK Runtime Environment (AdoptOpenJDK)(build 1.8.0_275-b01); OpenJDK 64-Bit Server VM (AdoptOpenJDK)(build 25.275-b01, mixed mode)
      Starting server from /Users/zkozick/dev/.virtualenvs/tqldev/lib/python3.7/site-packages/h2o/backend/bin/h2o.jar
      Ice root: /var/folders/2_/tx9l8xzd4fxg22hdd2wgd7d40000gn/T/tmpt2nmth_3
      JVM stdout: /var/folders/2_/tx9l8xzd4fxg22hdd2wgd7d40000gn/T/tmpt2nmth_3/h2o_zkozick_started_from_python.out
      JVM stderr: /var/folders/2_/tx9l8xzd4fxg22hdd2wgd7d40000gn/T/tmpt2nmth_3/h2o_zkozick_started_from_python.err
      Server is running at http://127.0.0.1:54321
    Connecting to H2O server at http://127.0.0.1:54321 ... successful.
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    |   iter    |  target   |   alpha   |
    -------------------------------------
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  1        | -0.000645 | -9.237    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  2        | -0.000629 | -2.201    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  3        | -0.000639 | -5.616    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  4        | -0.000629 | -2.765    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  5        | -0.000629 | -0.2201   |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  6        | -0.000629 | -4.615    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  7        | -0.000629 | -4.989    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  8        | -0.000645 | -9.279    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  9        | -0.000644 | -7.316    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  10       | -0.000629 | -5.001    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  11       | -0.000629 | -1.149    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  12       | -0.000629 | -3.67     |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  13       | -0.000629 | -1.658    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  14       | -0.000629 | -0.6461   |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  15       | -0.000629 | -4.11     |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  16       | -0.000629 | -3.209    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  17       | -0.000629 | -0.000322 |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  18       | -0.000629 | -2.479    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  19       | -0.000629 | -1.405    |
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    |  20       | -0.000629 | -4.358    |
    =====================================
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    Parse progress: |█████████████████████████████████████████████████████████| 100%
    glm Model Build progress: |███████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%
    glm prediction progress: |████████████████████████████████████████████████| 100%



<div style="overflow:auto"><table style="width:50%"><tr><td>H2O_cluster_uptime:</td>
<td>02 secs</td></tr>
<tr><td>H2O_cluster_timezone:</td>
<td>America/Denver</td></tr>
<tr><td>H2O_data_parsing_timezone:</td>
<td>UTC</td></tr>
<tr><td>H2O_cluster_version:</td>
<td>3.32.1.6</td></tr>
<tr><td>H2O_cluster_version_age:</td>
<td>11 days </td></tr>
<tr><td>H2O_cluster_name:</td>
<td>H2O_from_python_zkozick_mqyqyc</td></tr>
<tr><td>H2O_cluster_total_nodes:</td>
<td>1</td></tr>
<tr><td>H2O_cluster_free_memory:</td>
<td>3.556 Gb</td></tr>
<tr><td>H2O_cluster_total_cores:</td>
<td>8</td></tr>
<tr><td>H2O_cluster_allowed_cores:</td>
<td>8</td></tr>
<tr><td>H2O_cluster_status:</td>
<td>accepting new members, healthy</td></tr>
<tr><td>H2O_connection_url:</td>
<td>http://127.0.0.1:54321</td></tr>
<tr><td>H2O_connection_proxy:</td>
<td>{"http": null, "https": null}</td></tr>
<tr><td>H2O_internal_security:</td>
<td>False</td></tr>
<tr><td>H2O_API_Extensions:</td>
<td>Amazon S3, XGBoost, Algos, AutoML, Core V3, TargetEncoder, Core V4</td></tr>
<tr><td>Python_version:</td>
<td>3.7.3 final</td></tr></table></div>


    Warning! Using fewer than 20 bootstraps can lead standard error estimates to vary by +/- 50%.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.
    /Users/zkozick/dev/nanigans/noumena/python/zeenk-tql/zeenk/tql/modeling/h2o/utilities.py:381: RuntimeWarning: invalid value encountered in true_divide
      incr_share = incr_pred / pred
    AUCc_realized calculation failed. Setting to 0.5.


## Summarize the Model Training Session


```python
model.summarize().coefficients()
```




<html>
  <head><style type="text/css">
#T_4e48e_ th {
  text-align: left;
}
#T_4e48e_row0_col0, #T_4e48e_row0_col1, #T_4e48e_row0_col2, #T_4e48e_row0_col3, #T_4e48e_row1_col0, #T_4e48e_row1_col1, #T_4e48e_row1_col2, #T_4e48e_row1_col3, #T_4e48e_row2_col0, #T_4e48e_row2_col1, #T_4e48e_row2_col2, #T_4e48e_row2_col3 {
  text-align: left;
  white-space: pre-wrap;
}
</style>
</head>
  <body><table style="display:inline; padding-right: 20px" id="T_4e48e_" class="tex2jax_ignore">
  <caption style="font-weight: bold; font-size:120%">GLM Coefficients</caption>
  <thead>
    <tr>
      <th class="col_heading level0 col0"/>
      <th class="col_heading level0 col1">Coefficient</th>
      <th class="col_heading level0 col2">Standard Error</th>
      <th class="col_heading level0 col3">t-Stat</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td id="T_4e48e_row0_col0" class="data row0 col0"><b>Intercept</b></td>
      <td id="T_4e48e_row0_col1" class="data row0 col1">0.712494</td>
      <td id="T_4e48e_row0_col2" class="data row0 col2">0.000628</td>
      <td id="T_4e48e_row0_col3" class="data row0 col3">1134.43</td>
    </tr>
    <tr>
      <td id="T_4e48e_row1_col0" class="data row1 col0"><b>ad_size.big</b></td>
      <td id="T_4e48e_row1_col1" class="data row1 col1">0.000000</td>
      <td id="T_4e48e_row1_col2" class="data row1 col2">0.000730</td>
      <td id="T_4e48e_row1_col3" class="data row1 col3">0.00</td>
    </tr>
    <tr>
      <td id="T_4e48e_row2_col0" class="data row2 col0"><b>ad_size.small</b></td>
      <td id="T_4e48e_row2_col1" class="data row2 col1">0.000000</td>
      <td id="T_4e48e_row2_col2" class="data row2 col2">0.000719</td>
      <td id="T_4e48e_row2_col3" class="data row2 col3">0.00</td>
    </tr>
  </tbody>
</table>
</body>
</html>





```python
model.summarize().model_metrics()
```




<html>
  <head><style type="text/css">
#T_fad97_ th {
  text-align: left;
}
#T_fad97_row0_col0, #T_fad97_row0_col1, #T_fad97_row0_col2, #T_fad97_row0_col3, #T_fad97_row0_col4, #T_fad97_row0_col5, #T_fad97_row0_col6, #T_fad97_row0_col7, #T_fad97_row0_col8, #T_fad97_row0_col9, #T_fad97_row0_col10, #T_fad97_row0_col11, #T_fad97_row0_col12, #T_fad97_row0_col13, #T_fad97_row0_col14, #T_fad97_row0_col15, #T_fad97_row1_col0, #T_fad97_row1_col1, #T_fad97_row1_col2, #T_fad97_row1_col3, #T_fad97_row1_col4, #T_fad97_row1_col5, #T_fad97_row1_col6, #T_fad97_row1_col7, #T_fad97_row1_col8, #T_fad97_row1_col9, #T_fad97_row1_col10, #T_fad97_row1_col11, #T_fad97_row1_col12, #T_fad97_row1_col13, #T_fad97_row1_col14, #T_fad97_row1_col15, #T_fad97_row2_col0, #T_fad97_row2_col1, #T_fad97_row2_col2, #T_fad97_row2_col3, #T_fad97_row2_col4, #T_fad97_row2_col5, #T_fad97_row2_col6, #T_fad97_row2_col7, #T_fad97_row2_col8, #T_fad97_row2_col9, #T_fad97_row2_col10, #T_fad97_row2_col11, #T_fad97_row2_col12, #T_fad97_row2_col13, #T_fad97_row2_col14, #T_fad97_row2_col15, #T_fad97_row3_col0, #T_fad97_row3_col1, #T_fad97_row3_col2, #T_fad97_row3_col3, #T_fad97_row3_col4, #T_fad97_row3_col5, #T_fad97_row3_col6, #T_fad97_row3_col7, #T_fad97_row3_col8, #T_fad97_row3_col9, #T_fad97_row3_col10, #T_fad97_row3_col11, #T_fad97_row3_col12, #T_fad97_row3_col13, #T_fad97_row3_col14, #T_fad97_row3_col15, #T_fad97_row4_col0, #T_fad97_row4_col1, #T_fad97_row4_col2, #T_fad97_row4_col3, #T_fad97_row4_col4, #T_fad97_row4_col5, #T_fad97_row4_col6, #T_fad97_row4_col7, #T_fad97_row4_col8, #T_fad97_row4_col9, #T_fad97_row4_col10, #T_fad97_row4_col11, #T_fad97_row4_col12, #T_fad97_row4_col13, #T_fad97_row4_col14, #T_fad97_row4_col15, #T_fad97_row5_col0, #T_fad97_row5_col1, #T_fad97_row5_col2, #T_fad97_row5_col3, #T_fad97_row5_col4, #T_fad97_row5_col5, #T_fad97_row5_col6, #T_fad97_row5_col7, #T_fad97_row5_col8, #T_fad97_row5_col9, #T_fad97_row5_col10, #T_fad97_row5_col11, #T_fad97_row5_col12, #T_fad97_row5_col13, #T_fad97_row5_col14, #T_fad97_row5_col15, #T_fad97_row6_col0, #T_fad97_row6_col1, #T_fad97_row6_col2, #T_fad97_row6_col3, #T_fad97_row6_col4, #T_fad97_row6_col5, #T_fad97_row6_col6, #T_fad97_row6_col7, #T_fad97_row6_col8, #T_fad97_row6_col9, #T_fad97_row6_col10, #T_fad97_row6_col11, #T_fad97_row6_col12, #T_fad97_row6_col13, #T_fad97_row6_col14, #T_fad97_row6_col15, #T_fad97_row7_col0, #T_fad97_row7_col1, #T_fad97_row7_col2, #T_fad97_row7_col3, #T_fad97_row7_col4, #T_fad97_row7_col5, #T_fad97_row7_col6, #T_fad97_row7_col7, #T_fad97_row7_col8, #T_fad97_row7_col9, #T_fad97_row7_col10, #T_fad97_row7_col11, #T_fad97_row7_col12, #T_fad97_row7_col13, #T_fad97_row7_col14, #T_fad97_row7_col15, #T_fad97_row8_col0, #T_fad97_row8_col1, #T_fad97_row8_col2, #T_fad97_row8_col3, #T_fad97_row8_col4, #T_fad97_row8_col5, #T_fad97_row8_col6, #T_fad97_row8_col7, #T_fad97_row8_col8, #T_fad97_row8_col9, #T_fad97_row8_col10, #T_fad97_row8_col11, #T_fad97_row8_col12, #T_fad97_row8_col13, #T_fad97_row8_col14, #T_fad97_row8_col15, #T_fad97_row9_col0, #T_fad97_row9_col1, #T_fad97_row9_col2, #T_fad97_row9_col3, #T_fad97_row9_col4, #T_fad97_row9_col5, #T_fad97_row9_col6, #T_fad97_row9_col7, #T_fad97_row9_col8, #T_fad97_row9_col9, #T_fad97_row9_col10, #T_fad97_row9_col11, #T_fad97_row9_col12, #T_fad97_row9_col13, #T_fad97_row9_col14, #T_fad97_row9_col15, #T_fad97_row10_col0, #T_fad97_row10_col1, #T_fad97_row10_col2, #T_fad97_row10_col3, #T_fad97_row10_col4, #T_fad97_row10_col5, #T_fad97_row10_col6, #T_fad97_row10_col7, #T_fad97_row10_col8, #T_fad97_row10_col9, #T_fad97_row10_col10, #T_fad97_row10_col11, #T_fad97_row10_col12, #T_fad97_row10_col13, #T_fad97_row10_col14, #T_fad97_row10_col15, #T_fad97_row11_col0, #T_fad97_row11_col1, #T_fad97_row11_col2, #T_fad97_row11_col3, #T_fad97_row11_col4, #T_fad97_row11_col5, #T_fad97_row11_col6, #T_fad97_row11_col7, #T_fad97_row11_col8, #T_fad97_row11_col9, #T_fad97_row11_col10, #T_fad97_row11_col11, #T_fad97_row11_col12, #T_fad97_row11_col13, #T_fad97_row11_col14, #T_fad97_row11_col15, #T_fad97_row12_col0, #T_fad97_row12_col1, #T_fad97_row12_col2, #T_fad97_row12_col3, #T_fad97_row12_col4, #T_fad97_row12_col5, #T_fad97_row12_col6, #T_fad97_row12_col7, #T_fad97_row12_col8, #T_fad97_row12_col9, #T_fad97_row12_col10, #T_fad97_row12_col11, #T_fad97_row12_col12, #T_fad97_row12_col13, #T_fad97_row12_col14, #T_fad97_row12_col15, #T_fad97_row13_col0, #T_fad97_row13_col1, #T_fad97_row13_col2, #T_fad97_row13_col3, #T_fad97_row13_col4, #T_fad97_row13_col5, #T_fad97_row13_col6, #T_fad97_row13_col7, #T_fad97_row13_col8, #T_fad97_row13_col9, #T_fad97_row13_col10, #T_fad97_row13_col11, #T_fad97_row13_col12, #T_fad97_row13_col13, #T_fad97_row13_col14, #T_fad97_row13_col15, #T_fad97_row14_col0, #T_fad97_row14_col1, #T_fad97_row14_col2, #T_fad97_row14_col3, #T_fad97_row14_col4, #T_fad97_row14_col5, #T_fad97_row14_col6, #T_fad97_row14_col7, #T_fad97_row14_col8, #T_fad97_row14_col9, #T_fad97_row14_col10, #T_fad97_row14_col11, #T_fad97_row14_col12, #T_fad97_row14_col13, #T_fad97_row14_col14, #T_fad97_row14_col15, #T_fad97_row15_col0, #T_fad97_row15_col1, #T_fad97_row15_col2, #T_fad97_row15_col3, #T_fad97_row15_col4, #T_fad97_row15_col5, #T_fad97_row15_col6, #T_fad97_row15_col7, #T_fad97_row15_col8, #T_fad97_row15_col9, #T_fad97_row15_col10, #T_fad97_row15_col11, #T_fad97_row15_col12, #T_fad97_row15_col13, #T_fad97_row15_col14, #T_fad97_row15_col15, #T_fad97_row16_col0, #T_fad97_row16_col1, #T_fad97_row16_col2, #T_fad97_row16_col3, #T_fad97_row16_col4, #T_fad97_row16_col5, #T_fad97_row16_col6, #T_fad97_row16_col7, #T_fad97_row16_col8, #T_fad97_row16_col9, #T_fad97_row16_col10, #T_fad97_row16_col11, #T_fad97_row16_col12, #T_fad97_row16_col13, #T_fad97_row16_col14, #T_fad97_row16_col15, #T_fad97_row17_col0, #T_fad97_row17_col1, #T_fad97_row17_col2, #T_fad97_row17_col3, #T_fad97_row17_col4, #T_fad97_row17_col5, #T_fad97_row17_col6, #T_fad97_row17_col7, #T_fad97_row17_col8, #T_fad97_row17_col9, #T_fad97_row17_col10, #T_fad97_row17_col11, #T_fad97_row17_col12, #T_fad97_row17_col13, #T_fad97_row17_col14, #T_fad97_row17_col15, #T_fad97_row18_col0, #T_fad97_row18_col1, #T_fad97_row18_col2, #T_fad97_row18_col3, #T_fad97_row18_col4, #T_fad97_row18_col5, #T_fad97_row18_col6, #T_fad97_row18_col7, #T_fad97_row18_col8, #T_fad97_row18_col9, #T_fad97_row18_col10, #T_fad97_row18_col11, #T_fad97_row18_col12, #T_fad97_row18_col13, #T_fad97_row18_col14, #T_fad97_row18_col15, #T_fad97_row19_col0, #T_fad97_row19_col1, #T_fad97_row19_col2, #T_fad97_row19_col3, #T_fad97_row19_col4, #T_fad97_row19_col5, #T_fad97_row19_col6, #T_fad97_row19_col7, #T_fad97_row19_col8, #T_fad97_row19_col9, #T_fad97_row19_col10, #T_fad97_row19_col11, #T_fad97_row19_col12, #T_fad97_row19_col13, #T_fad97_row19_col14, #T_fad97_row19_col15, #T_fad97_row20_col0, #T_fad97_row20_col1, #T_fad97_row20_col2, #T_fad97_row20_col3, #T_fad97_row20_col4, #T_fad97_row20_col5, #T_fad97_row20_col6, #T_fad97_row20_col7, #T_fad97_row20_col8, #T_fad97_row20_col9, #T_fad97_row20_col10, #T_fad97_row20_col11, #T_fad97_row20_col12, #T_fad97_row20_col13, #T_fad97_row20_col14, #T_fad97_row20_col15, #T_fad97_row21_col0, #T_fad97_row21_col1, #T_fad97_row21_col2, #T_fad97_row21_col3, #T_fad97_row21_col4, #T_fad97_row21_col5, #T_fad97_row21_col6, #T_fad97_row21_col7, #T_fad97_row21_col8, #T_fad97_row21_col9, #T_fad97_row21_col10, #T_fad97_row21_col11, #T_fad97_row21_col12, #T_fad97_row21_col13, #T_fad97_row21_col14, #T_fad97_row21_col15, #T_fad97_row22_col0, #T_fad97_row22_col1, #T_fad97_row22_col2, #T_fad97_row22_col3, #T_fad97_row22_col4, #T_fad97_row22_col5, #T_fad97_row22_col6, #T_fad97_row22_col7, #T_fad97_row22_col8, #T_fad97_row22_col9, #T_fad97_row22_col10, #T_fad97_row22_col11, #T_fad97_row22_col12, #T_fad97_row22_col13, #T_fad97_row22_col14, #T_fad97_row22_col15, #T_fad97_row23_col0, #T_fad97_row23_col1, #T_fad97_row23_col2, #T_fad97_row23_col3, #T_fad97_row23_col4, #T_fad97_row23_col5, #T_fad97_row23_col6, #T_fad97_row23_col7, #T_fad97_row23_col8, #T_fad97_row23_col9, #T_fad97_row23_col10, #T_fad97_row23_col11, #T_fad97_row23_col12, #T_fad97_row23_col13, #T_fad97_row23_col14, #T_fad97_row23_col15, #T_fad97_row24_col0, #T_fad97_row24_col1, #T_fad97_row24_col2, #T_fad97_row24_col3, #T_fad97_row24_col4, #T_fad97_row24_col5, #T_fad97_row24_col6, #T_fad97_row24_col7, #T_fad97_row24_col8, #T_fad97_row24_col9, #T_fad97_row24_col10, #T_fad97_row24_col11, #T_fad97_row24_col12, #T_fad97_row24_col13, #T_fad97_row24_col14, #T_fad97_row24_col15, #T_fad97_row25_col0, #T_fad97_row25_col1, #T_fad97_row25_col2, #T_fad97_row25_col3, #T_fad97_row25_col4, #T_fad97_row25_col5, #T_fad97_row25_col6, #T_fad97_row25_col7, #T_fad97_row25_col8, #T_fad97_row25_col9, #T_fad97_row25_col10, #T_fad97_row25_col11, #T_fad97_row25_col12, #T_fad97_row25_col13, #T_fad97_row25_col14, #T_fad97_row25_col15, #T_fad97_row26_col0, #T_fad97_row26_col1, #T_fad97_row26_col2, #T_fad97_row26_col3, #T_fad97_row26_col4, #T_fad97_row26_col5, #T_fad97_row26_col6, #T_fad97_row26_col7, #T_fad97_row26_col8, #T_fad97_row26_col9, #T_fad97_row26_col10, #T_fad97_row26_col11, #T_fad97_row26_col12, #T_fad97_row26_col13, #T_fad97_row26_col14, #T_fad97_row26_col15, #T_fad97_row27_col0, #T_fad97_row27_col1, #T_fad97_row27_col2, #T_fad97_row27_col3, #T_fad97_row27_col4, #T_fad97_row27_col5, #T_fad97_row27_col6, #T_fad97_row27_col7, #T_fad97_row27_col8, #T_fad97_row27_col9, #T_fad97_row27_col10, #T_fad97_row27_col11, #T_fad97_row27_col12, #T_fad97_row27_col13, #T_fad97_row27_col14, #T_fad97_row27_col15, #T_fad97_row28_col0, #T_fad97_row28_col1, #T_fad97_row28_col2, #T_fad97_row28_col3, #T_fad97_row28_col4, #T_fad97_row28_col5, #T_fad97_row28_col6, #T_fad97_row28_col7, #T_fad97_row28_col8, #T_fad97_row28_col9, #T_fad97_row28_col10, #T_fad97_row28_col11, #T_fad97_row28_col12, #T_fad97_row28_col13, #T_fad97_row28_col14, #T_fad97_row28_col15, #T_fad97_row29_col0, #T_fad97_row29_col1, #T_fad97_row29_col2, #T_fad97_row29_col3, #T_fad97_row29_col4, #T_fad97_row29_col5, #T_fad97_row29_col6, #T_fad97_row29_col7, #T_fad97_row29_col8, #T_fad97_row29_col9, #T_fad97_row29_col10, #T_fad97_row29_col11, #T_fad97_row29_col12, #T_fad97_row29_col13, #T_fad97_row29_col14, #T_fad97_row29_col15, #T_fad97_row30_col0, #T_fad97_row30_col1, #T_fad97_row30_col2, #T_fad97_row30_col3, #T_fad97_row30_col4, #T_fad97_row30_col5, #T_fad97_row30_col6, #T_fad97_row30_col7, #T_fad97_row30_col8, #T_fad97_row30_col9, #T_fad97_row30_col10, #T_fad97_row30_col11, #T_fad97_row30_col12, #T_fad97_row30_col13, #T_fad97_row30_col14, #T_fad97_row30_col15, #T_fad97_row31_col0, #T_fad97_row31_col1, #T_fad97_row31_col2, #T_fad97_row31_col3, #T_fad97_row31_col4, #T_fad97_row31_col5, #T_fad97_row31_col6, #T_fad97_row31_col7, #T_fad97_row31_col8, #T_fad97_row31_col9, #T_fad97_row31_col10, #T_fad97_row31_col11, #T_fad97_row31_col12, #T_fad97_row31_col13, #T_fad97_row31_col14, #T_fad97_row31_col15, #T_fad97_row32_col0, #T_fad97_row32_col1, #T_fad97_row32_col2, #T_fad97_row32_col3, #T_fad97_row32_col4, #T_fad97_row32_col5, #T_fad97_row32_col6, #T_fad97_row32_col7, #T_fad97_row32_col8, #T_fad97_row32_col9, #T_fad97_row32_col10, #T_fad97_row32_col11, #T_fad97_row32_col12, #T_fad97_row32_col13, #T_fad97_row32_col14, #T_fad97_row32_col15, #T_fad97_row33_col0, #T_fad97_row33_col1, #T_fad97_row33_col2, #T_fad97_row33_col3, #T_fad97_row33_col4, #T_fad97_row33_col5, #T_fad97_row33_col6, #T_fad97_row33_col7, #T_fad97_row33_col8, #T_fad97_row33_col9, #T_fad97_row33_col10, #T_fad97_row33_col11, #T_fad97_row33_col12, #T_fad97_row33_col13, #T_fad97_row33_col14, #T_fad97_row33_col15, #T_fad97_row34_col0, #T_fad97_row34_col1, #T_fad97_row34_col2, #T_fad97_row34_col3, #T_fad97_row34_col4, #T_fad97_row34_col5, #T_fad97_row34_col6, #T_fad97_row34_col7, #T_fad97_row34_col8, #T_fad97_row34_col9, #T_fad97_row34_col10, #T_fad97_row34_col11, #T_fad97_row34_col12, #T_fad97_row34_col13, #T_fad97_row34_col14, #T_fad97_row34_col15 {
  text-align: left;
  white-space: pre-wrap;
}
</style>
</head>
  <body><table style="display:inline; padding-right: 20px" id="T_fad97_" class="tex2jax_ignore">
  <caption style="font-weight: bold; font-size:120%">Model Metrics</caption>
  <thead>
    <tr>
      <th class="col_heading level0 col0"/>
      <th class="col_heading level0 col1">Value 
(Train+Test)</th>
      <th class="col_heading level0 col2">Standard Error 
(Train+Test)</th>
      <th class="col_heading level0 col3">t-stat 
(Train+Test)</th>
      <th class="col_heading level0 col4">Lower 95 
(Train+Test)</th>
      <th class="col_heading level0 col5">Upper 95 
(Train+Test)</th>
      <th class="col_heading level0 col6">Value 
(Train)</th>
      <th class="col_heading level0 col7">Standard Error 
(Train)</th>
      <th class="col_heading level0 col8">t-stat 
(Train)</th>
      <th class="col_heading level0 col9">Lower 95 
(Train)</th>
      <th class="col_heading level0 col10">Upper 95 
(Train)</th>
      <th class="col_heading level0 col11">Value 
(Test)</th>
      <th class="col_heading level0 col12">Standard Error 
(Test)</th>
      <th class="col_heading level0 col13">t-stat 
(Test)</th>
      <th class="col_heading level0 col14">Lower 95 
(Test)</th>
      <th class="col_heading level0 col15">Upper 95 
(Test)</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td id="T_fad97_row0_col0" class="data row0 col0"><b>r2</b></td>
      <td id="T_fad97_row0_col1" class="data row0 col1">-0.000024</td>
      <td id="T_fad97_row0_col2" class="data row0 col2">0.000013</td>
      <td id="T_fad97_row0_col3" class="data row0 col3">-1.82</td>
      <td id="T_fad97_row0_col4" class="data row0 col4">-0.000051</td>
      <td id="T_fad97_row0_col5" class="data row0 col5">0.000002</td>
      <td id="T_fad97_row0_col6" class="data row0 col6">0.000000</td>
      <td id="T_fad97_row0_col7" class="data row0 col7">0.000008</td>
      <td id="T_fad97_row0_col8" class="data row0 col8">0.00</td>
      <td id="T_fad97_row0_col9" class="data row0 col9">-0.000015</td>
      <td id="T_fad97_row0_col10" class="data row0 col10">0.000015</td>
      <td id="T_fad97_row0_col11" class="data row0 col11">-0.000630</td>
      <td id="T_fad97_row0_col12" class="data row0 col12">0.000231</td>
      <td id="T_fad97_row0_col13" class="data row0 col13">-2.73</td>
      <td id="T_fad97_row0_col14" class="data row0 col14">-0.001083</td>
      <td id="T_fad97_row0_col15" class="data row0 col15">-0.000177</td>
    </tr>
    <tr>
      <td id="T_fad97_row1_col0" class="data row1 col0"><b>r2_incr</b></td>
      <td id="T_fad97_row1_col1" class="data row1 col1">0.000000</td>
      <td id="T_fad97_row1_col2" class="data row1 col2">0.000000</td>
      <td id="T_fad97_row1_col3" class="data row1 col3">0.00</td>
      <td id="T_fad97_row1_col4" class="data row1 col4">0.000000</td>
      <td id="T_fad97_row1_col5" class="data row1 col5">0.000000</td>
      <td id="T_fad97_row1_col6" class="data row1 col6">0.000000</td>
      <td id="T_fad97_row1_col7" class="data row1 col7">0.000000</td>
      <td id="T_fad97_row1_col8" class="data row1 col8">0.00</td>
      <td id="T_fad97_row1_col9" class="data row1 col9">0.000000</td>
      <td id="T_fad97_row1_col10" class="data row1 col10">0.000000</td>
      <td id="T_fad97_row1_col11" class="data row1 col11">0.000000</td>
      <td id="T_fad97_row1_col12" class="data row1 col12">0.000000</td>
      <td id="T_fad97_row1_col13" class="data row1 col13">0.00</td>
      <td id="T_fad97_row1_col14" class="data row1 col14">0.000000</td>
      <td id="T_fad97_row1_col15" class="data row1 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row2_col0" class="data row2 col0"><b>r2_het</b></td>
      <td id="T_fad97_row2_col1" class="data row2 col1">0.000000</td>
      <td id="T_fad97_row2_col2" class="data row2 col2">0.000000</td>
      <td id="T_fad97_row2_col3" class="data row2 col3">0.00</td>
      <td id="T_fad97_row2_col4" class="data row2 col4">0.000000</td>
      <td id="T_fad97_row2_col5" class="data row2 col5">0.000000</td>
      <td id="T_fad97_row2_col6" class="data row2 col6">0.000000</td>
      <td id="T_fad97_row2_col7" class="data row2 col7">0.000000</td>
      <td id="T_fad97_row2_col8" class="data row2 col8">0.00</td>
      <td id="T_fad97_row2_col9" class="data row2 col9">0.000000</td>
      <td id="T_fad97_row2_col10" class="data row2 col10">0.000000</td>
      <td id="T_fad97_row2_col11" class="data row2 col11">0.000000</td>
      <td id="T_fad97_row2_col12" class="data row2 col12">0.000000</td>
      <td id="T_fad97_row2_col13" class="data row2 col13">0.00</td>
      <td id="T_fad97_row2_col14" class="data row2 col14">0.000000</td>
      <td id="T_fad97_row2_col15" class="data row2 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row3_col0" class="data row3 col0"><b>auc</b></td>
      <td id="T_fad97_row3_col1" class="data row3 col1">0.500000</td>
      <td id="T_fad97_row3_col2" class="data row3 col2">0.000927</td>
      <td id="T_fad97_row3_col3" class="data row3 col3">539.10</td>
      <td id="T_fad97_row3_col4" class="data row3 col4">0.498182</td>
      <td id="T_fad97_row3_col5" class="data row3 col5">0.501818</td>
      <td id="T_fad97_row3_col6" class="data row3 col6">0.500000</td>
      <td id="T_fad97_row3_col7" class="data row3 col7">0.000820</td>
      <td id="T_fad97_row3_col8" class="data row3 col8">610.04</td>
      <td id="T_fad97_row3_col9" class="data row3 col9">0.498394</td>
      <td id="T_fad97_row3_col10" class="data row3 col10">0.501606</td>
      <td id="T_fad97_row3_col11" class="data row3 col11">0.500000</td>
      <td id="T_fad97_row3_col12" class="data row3 col12">0.001569</td>
      <td id="T_fad97_row3_col13" class="data row3 col13">318.71</td>
      <td id="T_fad97_row3_col14" class="data row3 col14">0.496925</td>
      <td id="T_fad97_row3_col15" class="data row3 col15">0.503075</td>
    </tr>
    <tr>
      <td id="T_fad97_row4_col0" class="data row4 col0"><b>aucc</b></td>
      <td id="T_fad97_row4_col1" class="data row4 col1">0.500000</td>
      <td id="T_fad97_row4_col2" class="data row4 col2">0.000927</td>
      <td id="T_fad97_row4_col3" class="data row4 col3">539.10</td>
      <td id="T_fad97_row4_col4" class="data row4 col4">0.498182</td>
      <td id="T_fad97_row4_col5" class="data row4 col5">0.501818</td>
      <td id="T_fad97_row4_col6" class="data row4 col6">0.500000</td>
      <td id="T_fad97_row4_col7" class="data row4 col7">0.000820</td>
      <td id="T_fad97_row4_col8" class="data row4 col8">610.04</td>
      <td id="T_fad97_row4_col9" class="data row4 col9">0.498394</td>
      <td id="T_fad97_row4_col10" class="data row4 col10">0.501606</td>
      <td id="T_fad97_row4_col11" class="data row4 col11">0.500000</td>
      <td id="T_fad97_row4_col12" class="data row4 col12">0.001569</td>
      <td id="T_fad97_row4_col13" class="data row4 col13">318.71</td>
      <td id="T_fad97_row4_col14" class="data row4 col14">0.496925</td>
      <td id="T_fad97_row4_col15" class="data row4 col15">0.503075</td>
    </tr>
    <tr>
      <td id="T_fad97_row5_col0" class="data row5 col0"><b>auc_incr</b></td>
      <td id="T_fad97_row5_col1" class="data row5 col1">0.000000</td>
      <td id="T_fad97_row5_col2" class="data row5 col2">0.000000</td>
      <td id="T_fad97_row5_col3" class="data row5 col3">0.00</td>
      <td id="T_fad97_row5_col4" class="data row5 col4">0.000000</td>
      <td id="T_fad97_row5_col5" class="data row5 col5">0.000000</td>
      <td id="T_fad97_row5_col6" class="data row5 col6">0.000000</td>
      <td id="T_fad97_row5_col7" class="data row5 col7">0.000000</td>
      <td id="T_fad97_row5_col8" class="data row5 col8">0.00</td>
      <td id="T_fad97_row5_col9" class="data row5 col9">0.000000</td>
      <td id="T_fad97_row5_col10" class="data row5 col10">0.000000</td>
      <td id="T_fad97_row5_col11" class="data row5 col11">0.000000</td>
      <td id="T_fad97_row5_col12" class="data row5 col12">0.000000</td>
      <td id="T_fad97_row5_col13" class="data row5 col13">0.00</td>
      <td id="T_fad97_row5_col14" class="data row5 col14">0.000000</td>
      <td id="T_fad97_row5_col15" class="data row5 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row6_col0" class="data row6 col0"><b>aucc_incr</b></td>
      <td id="T_fad97_row6_col1" class="data row6 col1">0.000000</td>
      <td id="T_fad97_row6_col2" class="data row6 col2">0.000000</td>
      <td id="T_fad97_row6_col3" class="data row6 col3">0.00</td>
      <td id="T_fad97_row6_col4" class="data row6 col4">0.000000</td>
      <td id="T_fad97_row6_col5" class="data row6 col5">0.000000</td>
      <td id="T_fad97_row6_col6" class="data row6 col6">0.000000</td>
      <td id="T_fad97_row6_col7" class="data row6 col7">0.000000</td>
      <td id="T_fad97_row6_col8" class="data row6 col8">0.00</td>
      <td id="T_fad97_row6_col9" class="data row6 col9">0.000000</td>
      <td id="T_fad97_row6_col10" class="data row6 col10">0.000000</td>
      <td id="T_fad97_row6_col11" class="data row6 col11">0.000000</td>
      <td id="T_fad97_row6_col12" class="data row6 col12">0.000000</td>
      <td id="T_fad97_row6_col13" class="data row6 col13">0.00</td>
      <td id="T_fad97_row6_col14" class="data row6 col14">0.000000</td>
      <td id="T_fad97_row6_col15" class="data row6 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row7_col0" class="data row7 col0"><b>aucc_realized</b></td>
      <td id="T_fad97_row7_col1" class="data row7 col1">0.500000</td>
      <td id="T_fad97_row7_col2" class="data row7 col2">0.000000</td>
      <td id="T_fad97_row7_col3" class="data row7 col3">0.00</td>
      <td id="T_fad97_row7_col4" class="data row7 col4">0.500000</td>
      <td id="T_fad97_row7_col5" class="data row7 col5">0.500000</td>
      <td id="T_fad97_row7_col6" class="data row7 col6">0.500000</td>
      <td id="T_fad97_row7_col7" class="data row7 col7">0.000000</td>
      <td id="T_fad97_row7_col8" class="data row7 col8">0.00</td>
      <td id="T_fad97_row7_col9" class="data row7 col9">0.500000</td>
      <td id="T_fad97_row7_col10" class="data row7 col10">0.500000</td>
      <td id="T_fad97_row7_col11" class="data row7 col11">0.500000</td>
      <td id="T_fad97_row7_col12" class="data row7 col12">0.000000</td>
      <td id="T_fad97_row7_col13" class="data row7 col13">0.00</td>
      <td id="T_fad97_row7_col14" class="data row7 col14">0.500000</td>
      <td id="T_fad97_row7_col15" class="data row7 col15">0.500000</td>
    </tr>
    <tr>
      <td id="T_fad97_row8_col0" class="data row8 col0"><b>auc_het</b></td>
      <td id="T_fad97_row8_col1" class="data row8 col1">0.000000</td>
      <td id="T_fad97_row8_col2" class="data row8 col2">0.000000</td>
      <td id="T_fad97_row8_col3" class="data row8 col3">0.00</td>
      <td id="T_fad97_row8_col4" class="data row8 col4">0.000000</td>
      <td id="T_fad97_row8_col5" class="data row8 col5">0.000000</td>
      <td id="T_fad97_row8_col6" class="data row8 col6">0.000000</td>
      <td id="T_fad97_row8_col7" class="data row8 col7">0.000000</td>
      <td id="T_fad97_row8_col8" class="data row8 col8">0.00</td>
      <td id="T_fad97_row8_col9" class="data row8 col9">0.000000</td>
      <td id="T_fad97_row8_col10" class="data row8 col10">0.000000</td>
      <td id="T_fad97_row8_col11" class="data row8 col11">0.000000</td>
      <td id="T_fad97_row8_col12" class="data row8 col12">0.000000</td>
      <td id="T_fad97_row8_col13" class="data row8 col13">0.00</td>
      <td id="T_fad97_row8_col14" class="data row8 col14">0.000000</td>
      <td id="T_fad97_row8_col15" class="data row8 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row9_col0" class="data row9 col0"><b>aucc_het</b></td>
      <td id="T_fad97_row9_col1" class="data row9 col1">0.000000</td>
      <td id="T_fad97_row9_col2" class="data row9 col2">0.000000</td>
      <td id="T_fad97_row9_col3" class="data row9 col3">0.00</td>
      <td id="T_fad97_row9_col4" class="data row9 col4">0.000000</td>
      <td id="T_fad97_row9_col5" class="data row9 col5">0.000000</td>
      <td id="T_fad97_row9_col6" class="data row9 col6">0.000000</td>
      <td id="T_fad97_row9_col7" class="data row9 col7">0.000000</td>
      <td id="T_fad97_row9_col8" class="data row9 col8">0.00</td>
      <td id="T_fad97_row9_col9" class="data row9 col9">0.000000</td>
      <td id="T_fad97_row9_col10" class="data row9 col10">0.000000</td>
      <td id="T_fad97_row9_col11" class="data row9 col11">0.000000</td>
      <td id="T_fad97_row9_col12" class="data row9 col12">0.000000</td>
      <td id="T_fad97_row9_col13" class="data row9 col13">0.00</td>
      <td id="T_fad97_row9_col14" class="data row9 col14">0.000000</td>
      <td id="T_fad97_row9_col15" class="data row9 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row10_col0" class="data row10 col0"><b>label_avg</b></td>
      <td id="T_fad97_row10_col1" class="data row10 col1">0.710252</td>
      <td id="T_fad97_row10_col2" class="data row10 col2">0.000604</td>
      <td id="T_fad97_row10_col3" class="data row10 col3">1175.83</td>
      <td id="T_fad97_row10_col4" class="data row10 col4">0.709068</td>
      <td id="T_fad97_row10_col5" class="data row10 col5">0.711436</td>
      <td id="T_fad97_row10_col6" class="data row10 col6">0.712494</td>
      <td id="T_fad97_row10_col7" class="data row10 col7">0.000628</td>
      <td id="T_fad97_row10_col8" class="data row10 col8">1134.40</td>
      <td id="T_fad97_row10_col9" class="data row10 col9">0.711263</td>
      <td id="T_fad97_row10_col10" class="data row10 col10">0.713725</td>
      <td id="T_fad97_row10_col11" class="data row10 col11">0.701004</td>
      <td id="T_fad97_row10_col12" class="data row10 col12">0.001918</td>
      <td id="T_fad97_row10_col13" class="data row10 col13">365.47</td>
      <td id="T_fad97_row10_col14" class="data row10 col14">0.697245</td>
      <td id="T_fad97_row10_col15" class="data row10 col15">0.704764</td>
    </tr>
    <tr>
      <td id="T_fad97_row11_col0" class="data row11 col0"><b>pred_avg</b></td>
      <td id="T_fad97_row11_col1" class="data row11 col1">0.712494</td>
      <td id="T_fad97_row11_col2" class="data row11 col2">0.000628</td>
      <td id="T_fad97_row11_col3" class="data row11 col3">1134.31</td>
      <td id="T_fad97_row11_col4" class="data row11 col4">0.711263</td>
      <td id="T_fad97_row11_col5" class="data row11 col5">0.713725</td>
      <td id="T_fad97_row11_col6" class="data row11 col6">0.712494</td>
      <td id="T_fad97_row11_col7" class="data row11 col7">0.000628</td>
      <td id="T_fad97_row11_col8" class="data row11 col8">1134.40</td>
      <td id="T_fad97_row11_col9" class="data row11 col9">0.711263</td>
      <td id="T_fad97_row11_col10" class="data row11 col10">0.713725</td>
      <td id="T_fad97_row11_col11" class="data row11 col11">0.712494</td>
      <td id="T_fad97_row11_col12" class="data row11 col12">0.000628</td>
      <td id="T_fad97_row11_col13" class="data row11 col13">1133.94</td>
      <td id="T_fad97_row11_col14" class="data row11 col14">0.711263</td>
      <td id="T_fad97_row11_col15" class="data row11 col15">0.713726</td>
    </tr>
    <tr>
      <td id="T_fad97_row12_col0" class="data row12 col0"><b>avg_error</b></td>
      <td id="T_fad97_row12_col1" class="data row12 col1">-0.002242</td>
      <td id="T_fad97_row12_col2" class="data row12 col2">0.000407</td>
      <td id="T_fad97_row12_col3" class="data row12 col3">-5.51</td>
      <td id="T_fad97_row12_col4" class="data row12 col4">-0.003040</td>
      <td id="T_fad97_row12_col5" class="data row12 col5">-0.001444</td>
      <td id="T_fad97_row12_col6" class="data row12 col6">0.000000</td>
      <td id="T_fad97_row12_col7" class="data row12 col7">0.000000</td>
      <td id="T_fad97_row12_col8" class="data row12 col8">0.86</td>
      <td id="T_fad97_row12_col9" class="data row12 col9">-0.000000</td>
      <td id="T_fad97_row12_col10" class="data row12 col10">0.000000</td>
      <td id="T_fad97_row12_col11" class="data row12 col11">-0.011490</td>
      <td id="T_fad97_row12_col12" class="data row12 col12">0.002071</td>
      <td id="T_fad97_row12_col13" class="data row12 col13">-5.55</td>
      <td id="T_fad97_row12_col14" class="data row12 col14">-0.015549</td>
      <td id="T_fad97_row12_col15" class="data row12 col15">-0.007431</td>
    </tr>
    <tr>
      <td id="T_fad97_row13_col0" class="data row13 col0"><b>realized_avg</b></td>
      <td id="T_fad97_row13_col1" class="data row13 col1">0.000000</td>
      <td id="T_fad97_row13_col2" class="data row13 col2">nan</td>
      <td id="T_fad97_row13_col3" class="data row13 col3">nan</td>
      <td id="T_fad97_row13_col4" class="data row13 col4">nan</td>
      <td id="T_fad97_row13_col5" class="data row13 col5">nan</td>
      <td id="T_fad97_row13_col6" class="data row13 col6">0.000000</td>
      <td id="T_fad97_row13_col7" class="data row13 col7">nan</td>
      <td id="T_fad97_row13_col8" class="data row13 col8">nan</td>
      <td id="T_fad97_row13_col9" class="data row13 col9">nan</td>
      <td id="T_fad97_row13_col10" class="data row13 col10">nan</td>
      <td id="T_fad97_row13_col11" class="data row13 col11">0.000000</td>
      <td id="T_fad97_row13_col12" class="data row13 col12">nan</td>
      <td id="T_fad97_row13_col13" class="data row13 col13">nan</td>
      <td id="T_fad97_row13_col14" class="data row13 col14">nan</td>
      <td id="T_fad97_row13_col15" class="data row13 col15">nan</td>
    </tr>
    <tr>
      <td id="T_fad97_row14_col0" class="data row14 col0"><b>incr_rate_avg</b></td>
      <td id="T_fad97_row14_col1" class="data row14 col1">0.000000</td>
      <td id="T_fad97_row14_col2" class="data row14 col2">0.000000</td>
      <td id="T_fad97_row14_col3" class="data row14 col3">0.00</td>
      <td id="T_fad97_row14_col4" class="data row14 col4">0.000000</td>
      <td id="T_fad97_row14_col5" class="data row14 col5">0.000000</td>
      <td id="T_fad97_row14_col6" class="data row14 col6">0.000000</td>
      <td id="T_fad97_row14_col7" class="data row14 col7">0.000000</td>
      <td id="T_fad97_row14_col8" class="data row14 col8">0.00</td>
      <td id="T_fad97_row14_col9" class="data row14 col9">0.000000</td>
      <td id="T_fad97_row14_col10" class="data row14 col10">0.000000</td>
      <td id="T_fad97_row14_col11" class="data row14 col11">0.000000</td>
      <td id="T_fad97_row14_col12" class="data row14 col12">0.000000</td>
      <td id="T_fad97_row14_col13" class="data row14 col13">0.00</td>
      <td id="T_fad97_row14_col14" class="data row14 col14">0.000000</td>
      <td id="T_fad97_row14_col15" class="data row14 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row15_col0" class="data row15 col0"><b>incr_rate_var</b></td>
      <td id="T_fad97_row15_col1" class="data row15 col1">0.000000</td>
      <td id="T_fad97_row15_col2" class="data row15 col2">0.000000</td>
      <td id="T_fad97_row15_col3" class="data row15 col3">0.00</td>
      <td id="T_fad97_row15_col4" class="data row15 col4">0.000000</td>
      <td id="T_fad97_row15_col5" class="data row15 col5">0.000000</td>
      <td id="T_fad97_row15_col6" class="data row15 col6">0.000000</td>
      <td id="T_fad97_row15_col7" class="data row15 col7">0.000000</td>
      <td id="T_fad97_row15_col8" class="data row15 col8">0.00</td>
      <td id="T_fad97_row15_col9" class="data row15 col9">0.000000</td>
      <td id="T_fad97_row15_col10" class="data row15 col10">0.000000</td>
      <td id="T_fad97_row15_col11" class="data row15 col11">0.000000</td>
      <td id="T_fad97_row15_col12" class="data row15 col12">0.000000</td>
      <td id="T_fad97_row15_col13" class="data row15 col13">0.00</td>
      <td id="T_fad97_row15_col14" class="data row15 col14">0.000000</td>
      <td id="T_fad97_row15_col15" class="data row15 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row16_col0" class="data row16 col0"><b>label_sum</b></td>
      <td id="T_fad97_row16_col1" class="data row16 col1">122147.000000</td>
      <td id="T_fad97_row16_col2" class="data row16 col2">316.913695</td>
      <td id="T_fad97_row16_col3" class="data row16 col3">385.43</td>
      <td id="T_fad97_row16_col4" class="data row16 col4">121525.860572</td>
      <td id="T_fad97_row16_col5" class="data row16 col5">122768.139428</td>
      <td id="T_fad97_row16_col6" class="data row16 col6">98622.000000</td>
      <td id="T_fad97_row16_col7" class="data row16 col7">290.989622</td>
      <td id="T_fad97_row16_col8" class="data row16 col8">338.92</td>
      <td id="T_fad97_row16_col9" class="data row16 col9">98051.670821</td>
      <td id="T_fad97_row16_col10" class="data row16 col10">99192.329179</td>
      <td id="T_fad97_row16_col11" class="data row16 col11">23525.000000</td>
      <td id="T_fad97_row16_col12" class="data row16 col12">124.511887</td>
      <td id="T_fad97_row16_col13" class="data row16 col13">188.94</td>
      <td id="T_fad97_row16_col14" class="data row16 col14">23280.961186</td>
      <td id="T_fad97_row16_col15" class="data row16 col15">23769.038814</td>
    </tr>
    <tr>
      <td id="T_fad97_row17_col0" class="data row17 col0"><b>pred_sum</b></td>
      <td id="T_fad97_row17_col1" class="data row17 col1">122532.587481</td>
      <td id="T_fad97_row17_col2" class="data row17 col2">311.074534</td>
      <td id="T_fad97_row17_col3" class="data row17 col3">393.90</td>
      <td id="T_fad97_row17_col4" class="data row17 col4">121922.892599</td>
      <td id="T_fad97_row17_col5" class="data row17 col5">123142.282364</td>
      <td id="T_fad97_row17_col6" class="data row17 col6">98622.000000</td>
      <td id="T_fad97_row17_col7" class="data row17 col7">290.989622</td>
      <td id="T_fad97_row17_col8" class="data row17 col8">338.92</td>
      <td id="T_fad97_row17_col9" class="data row17 col9">98051.670821</td>
      <td id="T_fad97_row17_col10" class="data row17 col10">99192.329179</td>
      <td id="T_fad97_row17_col11" class="data row17 col11">23910.587481</td>
      <td id="T_fad97_row17_col12" class="data row17 col12">127.164609</td>
      <td id="T_fad97_row17_col13" class="data row17 col13">188.03</td>
      <td id="T_fad97_row17_col14" class="data row17 col14">23661.349429</td>
      <td id="T_fad97_row17_col15" class="data row17 col15">24159.825534</td>
    </tr>
    <tr>
      <td id="T_fad97_row18_col0" class="data row18 col0"><b>conv_rate</b></td>
      <td id="T_fad97_row18_col1" class="data row18 col1">122532.587481</td>
      <td id="T_fad97_row18_col2" class="data row18 col2">311.074534</td>
      <td id="T_fad97_row18_col3" class="data row18 col3">393.90</td>
      <td id="T_fad97_row18_col4" class="data row18 col4">121922.892599</td>
      <td id="T_fad97_row18_col5" class="data row18 col5">123142.282364</td>
      <td id="T_fad97_row18_col6" class="data row18 col6">98622.000000</td>
      <td id="T_fad97_row18_col7" class="data row18 col7">290.989622</td>
      <td id="T_fad97_row18_col8" class="data row18 col8">338.92</td>
      <td id="T_fad97_row18_col9" class="data row18 col9">98051.670821</td>
      <td id="T_fad97_row18_col10" class="data row18 col10">99192.329179</td>
      <td id="T_fad97_row18_col11" class="data row18 col11">23910.587481</td>
      <td id="T_fad97_row18_col12" class="data row18 col12">127.164609</td>
      <td id="T_fad97_row18_col13" class="data row18 col13">188.03</td>
      <td id="T_fad97_row18_col14" class="data row18 col14">23661.349429</td>
      <td id="T_fad97_row18_col15" class="data row18 col15">24159.825534</td>
    </tr>
    <tr>
      <td id="T_fad97_row19_col0" class="data row19 col0"><b>incr_rate</b></td>
      <td id="T_fad97_row19_col1" class="data row19 col1">0.000000</td>
      <td id="T_fad97_row19_col2" class="data row19 col2">0.000000</td>
      <td id="T_fad97_row19_col3" class="data row19 col3">0.00</td>
      <td id="T_fad97_row19_col4" class="data row19 col4">0.000000</td>
      <td id="T_fad97_row19_col5" class="data row19 col5">0.000000</td>
      <td id="T_fad97_row19_col6" class="data row19 col6">0.000000</td>
      <td id="T_fad97_row19_col7" class="data row19 col7">0.000000</td>
      <td id="T_fad97_row19_col8" class="data row19 col8">0.00</td>
      <td id="T_fad97_row19_col9" class="data row19 col9">0.000000</td>
      <td id="T_fad97_row19_col10" class="data row19 col10">0.000000</td>
      <td id="T_fad97_row19_col11" class="data row19 col11">0.000000</td>
      <td id="T_fad97_row19_col12" class="data row19 col12">0.000000</td>
      <td id="T_fad97_row19_col13" class="data row19 col13">0.00</td>
      <td id="T_fad97_row19_col14" class="data row19 col14">0.000000</td>
      <td id="T_fad97_row19_col15" class="data row19 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row20_col0" class="data row20 col0"><b>realized</b></td>
      <td id="T_fad97_row20_col1" class="data row20 col1">0.000000</td>
      <td id="T_fad97_row20_col2" class="data row20 col2">nan</td>
      <td id="T_fad97_row20_col3" class="data row20 col3">nan</td>
      <td id="T_fad97_row20_col4" class="data row20 col4">nan</td>
      <td id="T_fad97_row20_col5" class="data row20 col5">nan</td>
      <td id="T_fad97_row20_col6" class="data row20 col6">0.000000</td>
      <td id="T_fad97_row20_col7" class="data row20 col7">nan</td>
      <td id="T_fad97_row20_col8" class="data row20 col8">nan</td>
      <td id="T_fad97_row20_col9" class="data row20 col9">nan</td>
      <td id="T_fad97_row20_col10" class="data row20 col10">nan</td>
      <td id="T_fad97_row20_col11" class="data row20 col11">0.000000</td>
      <td id="T_fad97_row20_col12" class="data row20 col12">nan</td>
      <td id="T_fad97_row20_col13" class="data row20 col13">nan</td>
      <td id="T_fad97_row20_col14" class="data row20 col14">nan</td>
      <td id="T_fad97_row20_col15" class="data row20 col15">nan</td>
    </tr>
    <tr>
      <td id="T_fad97_row21_col0" class="data row21 col0"><b>realized_raw</b></td>
      <td id="T_fad97_row21_col1" class="data row21 col1">0.000000</td>
      <td id="T_fad97_row21_col2" class="data row21 col2">nan</td>
      <td id="T_fad97_row21_col3" class="data row21 col3">nan</td>
      <td id="T_fad97_row21_col4" class="data row21 col4">nan</td>
      <td id="T_fad97_row21_col5" class="data row21 col5">nan</td>
      <td id="T_fad97_row21_col6" class="data row21 col6">0.000000</td>
      <td id="T_fad97_row21_col7" class="data row21 col7">nan</td>
      <td id="T_fad97_row21_col8" class="data row21 col8">nan</td>
      <td id="T_fad97_row21_col9" class="data row21 col9">nan</td>
      <td id="T_fad97_row21_col10" class="data row21 col10">nan</td>
      <td id="T_fad97_row21_col11" class="data row21 col11">0.000000</td>
      <td id="T_fad97_row21_col12" class="data row21 col12">nan</td>
      <td id="T_fad97_row21_col13" class="data row21 col13">nan</td>
      <td id="T_fad97_row21_col14" class="data row21 col14">nan</td>
      <td id="T_fad97_row21_col15" class="data row21 col15">nan</td>
    </tr>
    <tr>
      <td id="T_fad97_row22_col0" class="data row22 col0"><b>het_rate</b></td>
      <td id="T_fad97_row22_col1" class="data row22 col1">0.000000</td>
      <td id="T_fad97_row22_col2" class="data row22 col2">0.000000</td>
      <td id="T_fad97_row22_col3" class="data row22 col3">0.00</td>
      <td id="T_fad97_row22_col4" class="data row22 col4">0.000000</td>
      <td id="T_fad97_row22_col5" class="data row22 col5">0.000000</td>
      <td id="T_fad97_row22_col6" class="data row22 col6">0.000000</td>
      <td id="T_fad97_row22_col7" class="data row22 col7">0.000000</td>
      <td id="T_fad97_row22_col8" class="data row22 col8">0.00</td>
      <td id="T_fad97_row22_col9" class="data row22 col9">0.000000</td>
      <td id="T_fad97_row22_col10" class="data row22 col10">0.000000</td>
      <td id="T_fad97_row22_col11" class="data row22 col11">0.000000</td>
      <td id="T_fad97_row22_col12" class="data row22 col12">0.000000</td>
      <td id="T_fad97_row22_col13" class="data row22 col13">0.00</td>
      <td id="T_fad97_row22_col14" class="data row22 col14">0.000000</td>
      <td id="T_fad97_row22_col15" class="data row22 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row23_col0" class="data row23 col0"><b>het_rate_abs</b></td>
      <td id="T_fad97_row23_col1" class="data row23 col1">0.000000</td>
      <td id="T_fad97_row23_col2" class="data row23 col2">0.000000</td>
      <td id="T_fad97_row23_col3" class="data row23 col3">0.00</td>
      <td id="T_fad97_row23_col4" class="data row23 col4">0.000000</td>
      <td id="T_fad97_row23_col5" class="data row23 col5">0.000000</td>
      <td id="T_fad97_row23_col6" class="data row23 col6">0.000000</td>
      <td id="T_fad97_row23_col7" class="data row23 col7">0.000000</td>
      <td id="T_fad97_row23_col8" class="data row23 col8">0.00</td>
      <td id="T_fad97_row23_col9" class="data row23 col9">0.000000</td>
      <td id="T_fad97_row23_col10" class="data row23 col10">0.000000</td>
      <td id="T_fad97_row23_col11" class="data row23 col11">0.000000</td>
      <td id="T_fad97_row23_col12" class="data row23 col12">0.000000</td>
      <td id="T_fad97_row23_col13" class="data row23 col13">0.00</td>
      <td id="T_fad97_row23_col14" class="data row23 col14">0.000000</td>
      <td id="T_fad97_row23_col15" class="data row23 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row24_col0" class="data row24 col0"><b>het_rate_sq</b></td>
      <td id="T_fad97_row24_col1" class="data row24 col1">0.000000</td>
      <td id="T_fad97_row24_col2" class="data row24 col2">0.000000</td>
      <td id="T_fad97_row24_col3" class="data row24 col3">0.00</td>
      <td id="T_fad97_row24_col4" class="data row24 col4">0.000000</td>
      <td id="T_fad97_row24_col5" class="data row24 col5">0.000000</td>
      <td id="T_fad97_row24_col6" class="data row24 col6">0.000000</td>
      <td id="T_fad97_row24_col7" class="data row24 col7">0.000000</td>
      <td id="T_fad97_row24_col8" class="data row24 col8">0.00</td>
      <td id="T_fad97_row24_col9" class="data row24 col9">0.000000</td>
      <td id="T_fad97_row24_col10" class="data row24 col10">0.000000</td>
      <td id="T_fad97_row24_col11" class="data row24 col11">0.000000</td>
      <td id="T_fad97_row24_col12" class="data row24 col12">0.000000</td>
      <td id="T_fad97_row24_col13" class="data row24 col13">0.00</td>
      <td id="T_fad97_row24_col14" class="data row24 col14">0.000000</td>
      <td id="T_fad97_row24_col15" class="data row24 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row25_col0" class="data row25 col0"><b>samples</b></td>
      <td id="T_fad97_row25_col1" class="data row25 col1">171977.000000</td>
      <td id="T_fad97_row25_col2" class="data row25 col2">0.000000</td>
      <td id="T_fad97_row25_col3" class="data row25 col3">0.00</td>
      <td id="T_fad97_row25_col4" class="data row25 col4">171977.000000</td>
      <td id="T_fad97_row25_col5" class="data row25 col5">171977.000000</td>
      <td id="T_fad97_row25_col6" class="data row25 col6">138418.000000</td>
      <td id="T_fad97_row25_col7" class="data row25 col7">0.000000</td>
      <td id="T_fad97_row25_col8" class="data row25 col8">0.00</td>
      <td id="T_fad97_row25_col9" class="data row25 col9">138418.000000</td>
      <td id="T_fad97_row25_col10" class="data row25 col10">138418.000000</td>
      <td id="T_fad97_row25_col11" class="data row25 col11">33559.000000</td>
      <td id="T_fad97_row25_col12" class="data row25 col12">0.000000</td>
      <td id="T_fad97_row25_col13" class="data row25 col13">0.00</td>
      <td id="T_fad97_row25_col14" class="data row25 col14">33559.000000</td>
      <td id="T_fad97_row25_col15" class="data row25 col15">33559.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row26_col0" class="data row26 col0"><b>weight</b></td>
      <td id="T_fad97_row26_col1" class="data row26 col1">171977.000000</td>
      <td id="T_fad97_row26_col2" class="data row26 col2">433.745190</td>
      <td id="T_fad97_row26_col3" class="data row26 col3">396.49</td>
      <td id="T_fad97_row26_col4" class="data row26 col4">171126.875049</td>
      <td id="T_fad97_row26_col5" class="data row26 col5">172827.124951</td>
      <td id="T_fad97_row26_col6" class="data row26 col6">138418.000000</td>
      <td id="T_fad97_row26_col7" class="data row26 col7">403.602094</td>
      <td id="T_fad97_row26_col8" class="data row26 col8">342.96</td>
      <td id="T_fad97_row26_col9" class="data row26 col9">137626.954432</td>
      <td id="T_fad97_row26_col10" class="data row26 col10">139209.045568</td>
      <td id="T_fad97_row26_col11" class="data row26 col11">33559.000000</td>
      <td id="T_fad97_row26_col12" class="data row26 col12">179.957884</td>
      <td id="T_fad97_row26_col13" class="data row26 col13">186.48</td>
      <td id="T_fad97_row26_col14" class="data row26 col14">33206.289029</td>
      <td id="T_fad97_row26_col15" class="data row26 col15">33911.710971</td>
    </tr>
    <tr>
      <td id="T_fad97_row27_col0" class="data row27 col0"><b>positives</b></td>
      <td id="T_fad97_row27_col1" class="data row27 col1">122147.000000</td>
      <td id="T_fad97_row27_col2" class="data row27 col2">0.000000</td>
      <td id="T_fad97_row27_col3" class="data row27 col3">0.00</td>
      <td id="T_fad97_row27_col4" class="data row27 col4">122147.000000</td>
      <td id="T_fad97_row27_col5" class="data row27 col5">122147.000000</td>
      <td id="T_fad97_row27_col6" class="data row27 col6">98622.000000</td>
      <td id="T_fad97_row27_col7" class="data row27 col7">0.000000</td>
      <td id="T_fad97_row27_col8" class="data row27 col8">0.00</td>
      <td id="T_fad97_row27_col9" class="data row27 col9">98622.000000</td>
      <td id="T_fad97_row27_col10" class="data row27 col10">98622.000000</td>
      <td id="T_fad97_row27_col11" class="data row27 col11">23525.000000</td>
      <td id="T_fad97_row27_col12" class="data row27 col12">0.000000</td>
      <td id="T_fad97_row27_col13" class="data row27 col13">0.00</td>
      <td id="T_fad97_row27_col14" class="data row27 col14">23525.000000</td>
      <td id="T_fad97_row27_col15" class="data row27 col15">23525.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row28_col0" class="data row28 col0"><b>negatives</b></td>
      <td id="T_fad97_row28_col1" class="data row28 col1">49830.000000</td>
      <td id="T_fad97_row28_col2" class="data row28 col2">0.000000</td>
      <td id="T_fad97_row28_col3" class="data row28 col3">0.00</td>
      <td id="T_fad97_row28_col4" class="data row28 col4">49830.000000</td>
      <td id="T_fad97_row28_col5" class="data row28 col5">49830.000000</td>
      <td id="T_fad97_row28_col6" class="data row28 col6">39796.000000</td>
      <td id="T_fad97_row28_col7" class="data row28 col7">0.000000</td>
      <td id="T_fad97_row28_col8" class="data row28 col8">0.00</td>
      <td id="T_fad97_row28_col9" class="data row28 col9">39796.000000</td>
      <td id="T_fad97_row28_col10" class="data row28 col10">39796.000000</td>
      <td id="T_fad97_row28_col11" class="data row28 col11">10034.000000</td>
      <td id="T_fad97_row28_col12" class="data row28 col12">0.000000</td>
      <td id="T_fad97_row28_col13" class="data row28 col13">0.00</td>
      <td id="T_fad97_row28_col14" class="data row28 col14">10034.000000</td>
      <td id="T_fad97_row28_col15" class="data row28 col15">10034.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row29_col0" class="data row29 col0"><b>rss</b></td>
      <td id="T_fad97_row29_col1" class="data row29 col1">35392.719304</td>
      <td id="T_fad97_row29_col2" class="data row29 col2">102.363402</td>
      <td id="T_fad97_row29_col3" class="data row29 col3">345.76</td>
      <td id="T_fad97_row29_col4" class="data row29 col4">35192.090722</td>
      <td id="T_fad97_row29_col5" class="data row29 col5">35593.347886</td>
      <td id="T_fad97_row29_col6" class="data row29 col6">28354.412808</td>
      <td id="T_fad97_row29_col7" class="data row29 col7">94.152122</td>
      <td id="T_fad97_row29_col8" class="data row29 col8">301.16</td>
      <td id="T_fad97_row29_col9" class="data row29 col9">28169.878039</td>
      <td id="T_fad97_row29_col10" class="data row29 col10">28538.947576</td>
      <td id="T_fad97_row29_col11" class="data row29 col11">7038.306496</td>
      <td id="T_fad97_row29_col12" class="data row29 col12">52.383240</td>
      <td id="T_fad97_row29_col13" class="data row29 col13">134.36</td>
      <td id="T_fad97_row29_col14" class="data row29 col14">6935.637232</td>
      <td id="T_fad97_row29_col15" class="data row29 col15">7140.975760</td>
    </tr>
    <tr>
      <td id="T_fad97_row30_col0" class="data row30 col0"><b>mse</b></td>
      <td id="T_fad97_row30_col1" class="data row30 col1">0.205799</td>
      <td id="T_fad97_row30_col2" class="data row30 col2">0.000254</td>
      <td id="T_fad97_row30_col3" class="data row30 col3">809.74</td>
      <td id="T_fad97_row30_col4" class="data row30 col4">0.205301</td>
      <td id="T_fad97_row30_col5" class="data row30 col5">0.206297</td>
      <td id="T_fad97_row30_col6" class="data row30 col6">0.204846</td>
      <td id="T_fad97_row30_col7" class="data row30 col7">0.000267</td>
      <td id="T_fad97_row30_col8" class="data row30 col8">768.02</td>
      <td id="T_fad97_row30_col9" class="data row30 col9">0.204324</td>
      <td id="T_fad97_row30_col10" class="data row30 col10">0.205369</td>
      <td id="T_fad97_row30_col11" class="data row30 col11">0.209729</td>
      <td id="T_fad97_row30_col12" class="data row30 col12">0.000816</td>
      <td id="T_fad97_row30_col13" class="data row30 col13">256.98</td>
      <td id="T_fad97_row30_col14" class="data row30 col14">0.208130</td>
      <td id="T_fad97_row30_col15" class="data row30 col15">0.211329</td>
    </tr>
    <tr>
      <td id="T_fad97_row31_col0" class="data row31 col0"><b>mse_base</b></td>
      <td id="T_fad97_row31_col1" class="data row31 col1">0.205799</td>
      <td id="T_fad97_row31_col2" class="data row31 col2">0.000254</td>
      <td id="T_fad97_row31_col3" class="data row31 col3">809.74</td>
      <td id="T_fad97_row31_col4" class="data row31 col4">0.205301</td>
      <td id="T_fad97_row31_col5" class="data row31 col5">0.206297</td>
      <td id="T_fad97_row31_col6" class="data row31 col6">0.204846</td>
      <td id="T_fad97_row31_col7" class="data row31 col7">0.000267</td>
      <td id="T_fad97_row31_col8" class="data row31 col8">768.02</td>
      <td id="T_fad97_row31_col9" class="data row31 col9">0.204324</td>
      <td id="T_fad97_row31_col10" class="data row31 col10">0.205369</td>
      <td id="T_fad97_row31_col11" class="data row31 col11">0.209729</td>
      <td id="T_fad97_row31_col12" class="data row31 col12">0.000816</td>
      <td id="T_fad97_row31_col13" class="data row31 col13">256.98</td>
      <td id="T_fad97_row31_col14" class="data row31 col14">0.208130</td>
      <td id="T_fad97_row31_col15" class="data row31 col15">0.211329</td>
    </tr>
    <tr>
      <td id="T_fad97_row32_col0" class="data row32 col0"><b>mse_hom</b></td>
      <td id="T_fad97_row32_col1" class="data row32 col1">0.205799</td>
      <td id="T_fad97_row32_col2" class="data row32 col2">0.000254</td>
      <td id="T_fad97_row32_col3" class="data row32 col3">809.74</td>
      <td id="T_fad97_row32_col4" class="data row32 col4">0.205301</td>
      <td id="T_fad97_row32_col5" class="data row32 col5">0.206297</td>
      <td id="T_fad97_row32_col6" class="data row32 col6">0.204846</td>
      <td id="T_fad97_row32_col7" class="data row32 col7">0.000267</td>
      <td id="T_fad97_row32_col8" class="data row32 col8">768.02</td>
      <td id="T_fad97_row32_col9" class="data row32 col9">0.204324</td>
      <td id="T_fad97_row32_col10" class="data row32 col10">0.205369</td>
      <td id="T_fad97_row32_col11" class="data row32 col11">0.209729</td>
      <td id="T_fad97_row32_col12" class="data row32 col12">0.000816</td>
      <td id="T_fad97_row32_col13" class="data row32 col13">256.98</td>
      <td id="T_fad97_row32_col14" class="data row32 col14">0.208130</td>
      <td id="T_fad97_row32_col15" class="data row32 col15">0.211329</td>
    </tr>
    <tr>
      <td id="T_fad97_row33_col0" class="data row33 col0"><b>mse_incr</b></td>
      <td id="T_fad97_row33_col1" class="data row33 col1">0.000000</td>
      <td id="T_fad97_row33_col2" class="data row33 col2">0.000000</td>
      <td id="T_fad97_row33_col3" class="data row33 col3">0.00</td>
      <td id="T_fad97_row33_col4" class="data row33 col4">0.000000</td>
      <td id="T_fad97_row33_col5" class="data row33 col5">0.000000</td>
      <td id="T_fad97_row33_col6" class="data row33 col6">0.000000</td>
      <td id="T_fad97_row33_col7" class="data row33 col7">0.000000</td>
      <td id="T_fad97_row33_col8" class="data row33 col8">0.00</td>
      <td id="T_fad97_row33_col9" class="data row33 col9">0.000000</td>
      <td id="T_fad97_row33_col10" class="data row33 col10">0.000000</td>
      <td id="T_fad97_row33_col11" class="data row33 col11">0.000000</td>
      <td id="T_fad97_row33_col12" class="data row33 col12">0.000000</td>
      <td id="T_fad97_row33_col13" class="data row33 col13">0.00</td>
      <td id="T_fad97_row33_col14" class="data row33 col14">0.000000</td>
      <td id="T_fad97_row33_col15" class="data row33 col15">0.000000</td>
    </tr>
    <tr>
      <td id="T_fad97_row34_col0" class="data row34 col0"><b>mse_het</b></td>
      <td id="T_fad97_row34_col1" class="data row34 col1">0.000000</td>
      <td id="T_fad97_row34_col2" class="data row34 col2">0.000000</td>
      <td id="T_fad97_row34_col3" class="data row34 col3">0.00</td>
      <td id="T_fad97_row34_col4" class="data row34 col4">0.000000</td>
      <td id="T_fad97_row34_col5" class="data row34 col5">0.000000</td>
      <td id="T_fad97_row34_col6" class="data row34 col6">0.000000</td>
      <td id="T_fad97_row34_col7" class="data row34 col7">0.000000</td>
      <td id="T_fad97_row34_col8" class="data row34 col8">0.00</td>
      <td id="T_fad97_row34_col9" class="data row34 col9">0.000000</td>
      <td id="T_fad97_row34_col10" class="data row34 col10">0.000000</td>
      <td id="T_fad97_row34_col11" class="data row34 col11">0.000000</td>
      <td id="T_fad97_row34_col12" class="data row34 col12">0.000000</td>
      <td id="T_fad97_row34_col13" class="data row34 col13">0.00</td>
      <td id="T_fad97_row34_col14" class="data row34 col14">0.000000</td>
      <td id="T_fad97_row34_col15" class="data row34 col15">0.000000</td>
    </tr>
  </tbody>
</table>
</body>
</html>




## Publish the Model into TQL:

The final step is to upload the trained model back into TQL so it may be used to make predictions.
To do that we use `model.publish()`, providing a name we would like to use to reference the model
in queries:


```python
model.publish('my_winrate_model')
```




    381



The `model.publish()` function returns the unique ID of the model upon successful publish.

## Make Model Predictions with TQL

Now that the model has been published into TQL, it can be used in conjunction with the special TQL
Expression function `PREDICT("model_type")` to make predictions against timeline events:


```python
select(
  col('bid.ad_size', 'ad_size'), 
  col('PREDICT("my_winrate_model")', 'predicting_winrate')
)\
.from_events('lethe4')\
.where('type == "bid"')\
.limit(3)
```




<h4><em>Query results:</em></h4>partition "_default"
<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 tex2jax_ignore">
  <thead>
    <tr style="text-align: right;">
      <th></th>
      <th>ad_size</th>
      <th>predicting_winrate</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>big</td>
      <td>0.7124940397925126</td>
    </tr>
    <tr>
      <th>1</th>
      <td>big</td>
      <td>0.7124940397925126</td>
    </tr>
    <tr>
      <th>2</th>
      <td>small</td>
      <td>0.7124940397925126</td>
    </tr>
  </tbody>
</table>
</div><i>query produced 3 rows x 2 columns in 0.43 seconds</i><br><br>



In this case, since the model was trained using features extracted from events of type `bid`, it
also only makes sense to make predictions on events of type `bid`.