# Using ResultSets

TQL ResultSet is the result of calling the [Query](queries) object's `submit()` method. ResultSets contain ***exactly one row per event***, and ***exactly one column per TQL expression*** defined in the `select()` clause. 

When submitting a [Query](queries), you can define it to be either interactive with `submit(interactive=True)` or
asynchronous with `submit(interative=False)`. Interactive queries store their ResultSet data in memory. Asynchronous
queries store their ResultSets on disk. ResultSets stored on disk can be recalled at a later time, see the
[load() function](#Load) below.


```python
from zeenk.tql import *
```

    ------------------------------------------------------------------
               Version: 20.1.17-SNAPSHOT
     Version Timestamp: 2021-08-27 20:56:17
           Version Age: 4 days, 13 hours, 39 minutes, 56 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: 1 days, 18 hours, 7 minutes, 55 seconds
    ------------------------------------------------------------------


## Loading ResultSet Data

### Loading Data as a Python List

The `.dataframe()` function returns the dataframe as a Python list of lists from all partitions:


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit()
# first row of the dataframe
resultset.dataframe()
```




    [[None, None, None, None, None, None, None],
     ['big', '1.0', '2021-01-02 00:02:36', 'false', 'r13260', 'u252', 'true'],
     ['big', '1.0', '2021-01-02 00:16:41', 'false', 'r13376', 'u252', 'true'],
     ['small', '1.0', '2021-01-02 00:30:55', 'false', 'r13509', 'u252', 'true'],
     [None, None, None, None, None, None, None]]



### Loading Data as a Pandas Dataframe
Similar to dataframe, `.pandas_dataframe()` returns the data, but as a [Pandas](https://pandas.pydata.org/) dataframe.


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit()
resultset.pandas_dataframe()
```




<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>
  </tbody>
</table>
</div>



### Loading Data as a Spark Dataframe
Similar to dataframe, `.spark_dataframe()` returns the data, but as a [Spark](https://spark.apache.org/docs/3.1.1/sql-programming-guide.html#untyped-dataset-operations-aka-dataframe-operations) dataframe. The backend Java process that manipulated the data outputs it as strings so Spark cannot always automatically infer data types.


```python
resultset = select("'test'", '1').from_events('lethe4').limit(5).submit()
spark_df = resultset.spark_dataframe()
spark_df.show(2)
```

    +----+---+
    |  _1| _2|
    +----+---+
    |test|  1|
    |test|  1|
    +----+---+
    only showing top 2 rows
    


## ResultSet Partitions
ResultSet data is always partitioned, though if the query did not define a `partition_key(expr)`, then there will
only be a single "default" partition with name `_default`.  If .`partition_key(expr)` was defined, then there will
be N partitions, one for each unique value output by the partition key expression.

### Default Partition
The method `default_partition()` outputs the contents of the default partition.


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit()
resultset.default_partition()
```




<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>
  </tbody>
</table>
</div>



### Partition Names
The `partition_names()` method outputs a tuple of partition names.


```python
resultset = select('activity.*')\
.from_events('lethe4')\
.limit(7)\
.partition_by('IF(RANDOM() < 0.5, "train", "test")')\
.submit()

resultset.partition_names()
```




    ('test', 'train')



### Partitions
The `partitions()` method outputs [Partition](tql.html#tql.resultset.Partition) objects.


```python
for part in resultset.partitions():
  print(f"""
name: {part.name()}
row_count: {part.row_count()}""")
```

    
    name: test
    row_count: 5
    
    name: train
    row_count: 2


### Partition
The `partition()` method gets the partition supplied as an argument. If no argument is supplied, it tries to get the default partition.


```python
resultset.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>activity_activity_id</th>
      <th>activity_conversion</th>
      <th>activity_event_time</th>
      <th>activity_user_id</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>1</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>2</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>3</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
    <tr>
      <th>4</th>
      <td>None</td>
      <td>None</td>
      <td>None</td>
      <td>None</td>
    </tr>
  </tbody>
</table>
</div>



## Retrieving Column Names
Calling column_names() retrieves the columns of the ResultSet.


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit()
resultset.column_names()

```

## Row Count
The `row_count()` method returns the total number of rows in the ResultSet over all partitions.


```python
resultset.row_count()
```




    7



## Positive row count
The `positive_row_count()` method returns the number of positive rows. 


```python
resultset.positive_row_count()
```




    0



## Get Query
The method `get_query()` returns the query that was used to generate the ResultSet


```python
query = resultset.get_query()
query.describe()
```




<style type="text/css">
#T_df2db_ th {
  text-align: left;
}
#T_df2db_row0_col0, #T_df2db_row0_col1, #T_df2db_row1_col0, #T_df2db_row1_col1, #T_df2db_row2_col0, #T_df2db_row2_col1, #T_df2db_row3_col0, #T_df2db_row3_col1, #T_df2db_row4_col0, #T_df2db_row4_col1, #T_df2db_row5_col0, #T_df2db_row5_col1 {
  text-align: left;
  white-space: pre-wrap;
}
</style>
<table style="display:inline; padding-right: 20px" id="T_df2db_">
  <caption style="font-weight: bold; font-size:120%">Query</caption style="font-weight: bold; font-size:120%">
  <thead>
    <tr>
      <th class="col_heading level0 col0" >Attribute</th>
      <th class="col_heading level0 col1" >Value</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td id="T_df2db_row0_col0" class="data row0 col0" >project_id</td>
      <td id="T_df2db_row0_col1" class="data row0 col1" >4</td>
    </tr>
    <tr>
      <td id="T_df2db_row1_col0" class="data row1 col0" >timeline_limit</td>
      <td id="T_df2db_row1_col1" class="data row1 col1" >-1</td>
    </tr>
    <tr>
      <td id="T_df2db_row2_col0" class="data row2 col0" >row_limit</td>
      <td id="T_df2db_row2_col1" class="data row2 col1" >7</td>
    </tr>
    <tr>
      <td id="T_df2db_row3_col0" class="data row3 col0" >partition_key_expression</td>
      <td id="T_df2db_row3_col1" class="data row3 col1" >IF(RANDOM() < 0.5, "train", "test")</td>
    </tr>
    <tr>
      <td id="T_df2db_row4_col0" class="data row4 col0" >interactive</td>
      <td id="T_df2db_row4_col1" class="data row4 col1" >True</td>
    </tr>
    <tr>
      <td id="T_df2db_row5_col0" class="data row5 col0" >allow_invalid_column_expressions</td>
      <td id="T_df2db_row5_col1" class="data row5 col1" >True</td>
    </tr>
  </tbody>
</table style="display:inline; padding-right: 20px">
<style type="text/css">
#T_f4dc5_ th {
  text-align: left;
}
#T_f4dc5_row0_col0, #T_f4dc5_row0_col1, #T_f4dc5_row0_col2, #T_f4dc5_row1_col0, #T_f4dc5_row1_col1, #T_f4dc5_row1_col2, #T_f4dc5_row2_col0, #T_f4dc5_row2_col1, #T_f4dc5_row2_col2, #T_f4dc5_row3_col0, #T_f4dc5_row3_col1, #T_f4dc5_row3_col2 {
  text-align: left;
  white-space: pre-wrap;
}
</style>
<table style="display:inline; padding-right: 20px" id="T_f4dc5_">
  <caption style="font-weight: bold; font-size:120%">Query.columns</caption style="font-weight: bold; font-size:120%">
  <thead>
    <tr>
      <th class="col_heading level0 col0" >name</th>
      <th class="col_heading level0 col1" >expression</th>
      <th class="col_heading level0 col2" >type</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td id="T_f4dc5_row0_col0" class="data row0 col0" >activity_activity_id</td>
      <td id="T_f4dc5_row0_col1" class="data row0 col1" >activity.activity_id</td>
      <td id="T_f4dc5_row0_col2" class="data row0 col2" >METADATA</td>
    </tr>
    <tr>
      <td id="T_f4dc5_row1_col0" class="data row1 col0" >activity_conversion</td>
      <td id="T_f4dc5_row1_col1" class="data row1 col1" >activity.conversion</td>
      <td id="T_f4dc5_row1_col2" class="data row1 col2" >METADATA</td>
    </tr>
    <tr>
      <td id="T_f4dc5_row2_col0" class="data row2 col0" >activity_event_time</td>
      <td id="T_f4dc5_row2_col1" class="data row2 col1" >activity.event_time</td>
      <td id="T_f4dc5_row2_col2" class="data row2 col2" >METADATA</td>
    </tr>
    <tr>
      <td id="T_f4dc5_row3_col0" class="data row3 col0" >activity_user_id</td>
      <td id="T_f4dc5_row3_col1" class="data row3 col1" >activity.user_id</td>
      <td id="T_f4dc5_row3_col2" class="data row3 col2" >METADATA</td>
    </tr>
  </tbody>
</table style="display:inline; padding-right: 20px">




## Get ID
The `get_id` method returns the ID of the ResultSet if the query was executed as a non-interactive job. Otherwise, it returns `-1`.


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit(interactive=False)
resultset.get_id()
```


Click to <a href="http://localhost:9000/v1/query/33/abort" target="_blank">Abort Job</a>
                            <p><em>NOTE: Abort link will open the result of the call in a <B>new tab/window</B></em></p>


    SUCCESS:  8/8|██████████|, TOTAL ELAPSED=0:00:17, WAIT=0:00:00, PENDING=0:00:01, RUNNING=0:00:16





    33



## Load
The `load()` method takes the ResultSet ID as an argument and loads it into the notebook. You can optionally set `analyze` parameter to `True` to get an analysis of the Query. 


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit(interactive=False)
rs_id = resultset.get_id()
load_resultset(rs_id).pandas_dataframe()
```


Click to <a href="http://localhost:9000/v1/query/34/abort" target="_blank">Abort Job</a>
                            <p><em>NOTE: Abort link will open the result of the call in a <B>new tab/window</B></em></p>


    SUCCESS:  8/8|██████████|, TOTAL ELAPSED=0:00:06, WAIT=0:00:00, PENDING=0:00:01, RUNNING=0:00:05





<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>
  </tbody>
</table>
</div>



## Refresh
The `refresh()` method refreshes non-interactive ResultSet fields after query execution finishes.


```python
resultset = select('bid.*').from_events('lethe4').limit(5).submit(interactive=False)
resultset.refresh()
resultset
```


Click to <a href="http://localhost:9000/v1/query/35/abort" target="_blank">Abort Job</a>
                            <p><em>NOTE: Abort link will open the result of the call in a <B>new tab/window</B></em></p>


    SUCCESS:  8/8|██████████|, TOTAL ELAPSED=0:00:07, WAIT=0:00:00, PENDING=0:00:01, RUNNING=0:00:06





<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>key</th>
      <th>value</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>name</td>
      <td>_default</td>
    </tr>
    <tr>
      <th>1</th>
      <td>data_path</td>
      <td>file:/Users/smaguire/.tql/files/datasets/2021-08-31/94096521418144/data/partition=_default</td>
    </tr>
    <tr>
      <th>2</th>
      <td>row_count</td>
      <td>40</td>
    </tr>
    <tr>
      <th>3</th>
      <td>columns</td>
      <td>[bid_ad_size, bid_bid, bid_event_time, bid_ghosted, bid_request_id, bid_user_id, bid_won]</td>
    </tr>
    <tr>
      <th>4</th>
      <td>positive_row_count</td>
      <td>0</td>
    </tr>
  </tbody>
</table>
</div><i>query produced 5 rows x 7 columns in 8.2 seconds</i><br><br>



## Metrics
The method `metrics()` returns timing and runtime metadata about the ResultSet.


```python
resultset = select(
  # column value metrics are computed when using FeatureColumn such as numerical() and categorical()
  numerical('1'),
  categorical('type', name='my_categorical_col')
).from_events('lethe4').limit(5).submit(analyze=True)
resultset.metrics()
```




<h4><em>Expression timing stats:</em></h4><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>name</th>
      <th>expression</th>
      <th>mean(ms)</th>
      <th>p95(ms)</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>_c0</td>
      <td>1</td>
      <td>0.007</td>
      <td>0.022</td>
    </tr>
    <tr>
      <th>1</th>
      <td>my_categorical_col</td>
      <td>type</td>
      <td>0.015</td>
      <td>0.046</td>
    </tr>
  </tbody>
</table>
</div><h4><em>Timeline & Event runtime:</em></h4><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>min</th>
      <th>max</th>
      <th>mean</th>
      <th>total</th>
      <th>p50</th>
      <th>p75</th>
      <th>p95</th>
      <th>p99</th>
      <th>unit</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>timeline_runtime</th>
      <td>7405</td>
      <td>7405</td>
      <td>7405</td>
      <td>7405</td>
      <td>7405.0</td>
      <td>7405.0</td>
      <td>7405.0</td>
      <td>7405.00</td>
      <td>micros</td>
    </tr>
    <tr>
      <th>event_runtime</th>
      <td>29</td>
      <td>652</td>
      <td>193</td>
      <td>773</td>
      <td>42.0</td>
      <td>50.0</td>
      <td>531.6</td>
      <td>627.92</td>
      <td>micros</td>
    </tr>
  </tbody>
</table>
</div><h4><em>Query summary:</em></h4><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>key</th>
      <th>value</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>query_runtime_micros</td>
      <td>21732</td>
    </tr>
    <tr>
      <th>1</th>
      <td>outcome_generated</td>
      <td>0</td>
    </tr>
    <tr>
      <th>2</th>
      <td>timelines_processed</td>
      <td>1</td>
    </tr>
    <tr>
      <th>3</th>
      <td>real_events_processed</td>
      <td>5</td>
    </tr>
    <tr>
      <th>4</th>
      <td>event_passed_filters</td>
      <td>5</td>
    </tr>
    <tr>
      <th>5</th>
      <td>opportunity_event_count</td>
      <td>0</td>
    </tr>
    <tr>
      <th>6</th>
      <td>null_label_evaluation_count</td>
      <td>0</td>
    </tr>
    <tr>
      <th>7</th>
      <td>null_weight_evaluation_count</td>
      <td>0</td>
    </tr>
    <tr>
      <th>8</th>
      <td>events_processed</td>
      <td>5</td>
    </tr>
    <tr>
      <th>9</th>
      <td>column_expanded</td>
      <td>False</td>
    </tr>
    <tr>
      <th>10</th>
      <td>dropped_empty_rows_after_conversion</td>
      <td>0</td>
    </tr>
    <tr>
      <th>11</th>
      <td>positive_example_count</td>
      <td>0</td>
    </tr>
    <tr>
      <th>12</th>
      <td>negative_example_count</td>
      <td>5</td>
    </tr>
    <tr>
      <th>13</th>
      <td>empty_example_count</td>
      <td>0</td>
    </tr>
    <tr>
      <th>14</th>
      <td>label_sum</td>
      <td>0.0</td>
    </tr>
    <tr>
      <th>15</th>
      <td>weighted_label_sum</td>
      <td>0.0</td>
    </tr>
  </tbody>
</table>
</div><h4><em>Column value metrics:</em></h4><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>name</th>
      <th>column_index</th>
      <th>positive_count</th>
      <th>negative_count</th>
      <th>non_null_count</th>
      <th>non_zero_count</th>
      <th>positive_sum</th>
      <th>negative_sum</th>
      <th>positive_weighted_count</th>
      <th>negative_weighted_count</th>
      <th>label_sum</th>
      <th>weighted_label_sum</th>
      <th>min_numerical_value</th>
      <th>max_numerical_value</th>
      <th>total_weighted_count</th>
      <th>total_sum</th>
      <th>total_count</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>_c0</td>
      <td>0</td>
      <td>0</td>
      <td>5</td>
      <td>5</td>
      <td>5</td>
      <td>0.0</td>
      <td>5.0</td>
      <td>0.0</td>
      <td>5.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>1.0</td>
      <td>1.0</td>
      <td>5.0</td>
      <td>5.0</td>
      <td>5</td>
    </tr>
    <tr>
      <th>1</th>
      <td>my_categorical_col_EQ_bid</td>
      <td>1</td>
      <td>0</td>
      <td>3</td>
      <td>3</td>
      <td>0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>3.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>NaN</td>
      <td>NaN</td>
      <td>3.0</td>
      <td>0.0</td>
      <td>3</td>
    </tr>
    <tr>
      <th>2</th>
      <td>my_categorical_col_EQ_activity</td>
      <td>1</td>
      <td>0</td>
      <td>1</td>
      <td>1</td>
      <td>0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>1.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>NaN</td>
      <td>NaN</td>
      <td>1.0</td>
      <td>0.0</td>
      <td>1</td>
    </tr>
    <tr>
      <th>3</th>
      <td>my_categorical_col_EQ_user</td>
      <td>1</td>
      <td>0</td>
      <td>1</td>
      <td>1</td>
      <td>0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>1.0</td>
      <td>0.0</td>
      <td>0.0</td>
      <td>NaN</td>
      <td>NaN</td>
      <td>1.0</td>
      <td>0.0</td>
      <td>1</td>
    </tr>
  </tbody>
</table>
</div>