Skip to content

Aggregation Strategies

Module for aggregation strategies.

Author: Daniel Wertheimer

AggregationStrategy

Bases: ABC

Abstract base class for aggregation strategies.

Subclasses should implement the aggregate method to perform a specific type of aggregation.

Methods:

Name Description
aggregate

Abstract method to be overridden by subclasses to perform aggregation.
include_col_name

Indicates whether to include the column name in the output.

Raises:

Type Description
NotImplementedError

If the method is not overridden by a subclass.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
class AggregationStrategy(ABC):
    """
    Abstract base class for aggregation strategies.

    Subclasses should implement the aggregate method to perform a specific type of aggregation.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Abstract method to be overridden by subclasses to perform aggregation.

    include_col_name() -> bool
        Indicates whether to include the column name in the output.

    Raises
    ------
    NotImplementedError
        If the method is not overridden by a subclass.
    """

    @abstractmethod
    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Perform aggregation on the specified column and group by the key columns.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        raise NotImplementedError("This method should be overridden by subclasses.")

    def include_col_name(self) -> bool:
        """
        Indicate whether to include the column name in the output.

        Returns
        -------
        bool
            True if the column name should be included, False otherwise.
        """
        return True

aggregate(df, agg_col, key_cols) abstractmethod

Perform aggregation on the specified column and group by the key columns.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
@abstractmethod
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Perform aggregation on the specified column and group by the key columns.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    raise NotImplementedError("This method should be overridden by subclasses.")

include_col_name()

Indicate whether to include the column name in the output.

Returns:

Type Description
bool

True if the column name should be included, False otherwise.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
55
56
57
58
59
60
61
62
63
64
def include_col_name(self) -> bool:
    """
    Indicate whether to include the column name in the output.

    Returns
    -------
    bool
        True if the column name should be included, False otherwise.
    """
    return True

CountAggregation

Bases: AggregationStrategy

CountAggregation class for counting occurrences of values in a specified column.

Attributes:

Name Type Description
include_missing bool

If True, missing values (nulls and NaNs) are included in the count aggregation. If False, missing values are excluded from the count.

Methods:

Name Description
aggregate

Aggregate data by counting occurrences.
_check_contains_missing

Checks if the specified column contains any missing values (nulls or NaNs).
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
class CountAggregation(AggregationStrategy):
    """CountAggregation class for counting occurrences of values in a specified column.

    Attributes
    ----------
    include_missing : bool
        If True, missing values (nulls and NaNs) are included in the count aggregation.
        If False, missing values are excluded from the count.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Aggregate data by counting occurrences.
    _check_contains_missing(df: DataFrame, agg_col: str) -> bool
        Checks if the specified column contains any missing values (nulls or NaNs).
    """

    def __init__(self, include_missing: Optional[bool] = False) -> None:
        """
        Initialise the CountAggregation class with the option to include or exclude missing values.

        Parameters
        ----------
        include_missing : bool, optional
            If True, include missing values in the count (default is False).

        Returns
        -------
        None
        """
        super().__init__()
        self.include_missing = include_missing

    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Aggregate data by counting occurrences.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        if self._check_contains_missing(df, agg_col):
            logger.warning(
                f"Column {agg_col} contains nulls. "
                f"Count aggregation may include null values. "
                f"If this is intentional, ignore this warning."
            )
        if not self.include_missing:
            output_df = df.dropna(subset=[agg_col]).groupby(*key_cols).agg(F.count(F.col(agg_col)))
        else:
            output_df = df.groupby(*key_cols).agg(F.count("*"))
        return output_df

    def _check_contains_missing(self, df: DataFrame, agg_col: str) -> bool:
        """
        Check if the specified column contains any missing values (nulls or NaNs).

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be checked.
        agg_col : str
            The column in the DataFrame to check for missing values.

        Returns
        -------
        bool
            True if the column contains missing values (nulls or NaNs), False otherwise.
        """
        null_count = df.filter(F.col(agg_col).isNull()).count()
        nan_count = df.filter(F.isnan(F.col(agg_col))).count()

        return null_count != 0 or nan_count != 0

aggregate(df, agg_col, key_cols)

Aggregate data by counting occurrences.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Aggregate data by counting occurrences.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    if self._check_contains_missing(df, agg_col):
        logger.warning(
            f"Column {agg_col} contains nulls. "
            f"Count aggregation may include null values. "
            f"If this is intentional, ignore this warning."
        )
    if not self.include_missing:
        output_df = df.dropna(subset=[agg_col]).groupby(*key_cols).agg(F.count(F.col(agg_col)))
    else:
        output_df = df.groupby(*key_cols).agg(F.count("*"))
    return output_df

CountIfOneAggregation

Bases: AggregationStrategy

CountIfOneAggregation class for counting the number of values that are equal to one.

Methods:

Name Description
aggregate

Aggregate data by counting occurrences of the number 1.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
class CountIfOneAggregation(AggregationStrategy):
    """CountIfOneAggregation class for counting the number of values that are equal to one.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Aggregate data by counting occurrences of the number 1.
    """

    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Aggregate data by counting occurrences of the number 1.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        return df.filter(F.col(agg_col) == 1).groupby(*key_cols).agg(F.count(F.col(agg_col)))

aggregate(df, agg_col, key_cols)

Aggregate data by counting occurrences of the number 1.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Aggregate data by counting occurrences of the number 1.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    return df.filter(F.col(agg_col) == 1).groupby(*key_cols).agg(F.count(F.col(agg_col)))

MeanAggregation

Bases: AggregationStrategy

MeanAggregation class for calculating the mean of values in a specified column.

Methods:

Name Description
aggregate

Aggregate data by calculating the mean.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
class MeanAggregation(AggregationStrategy):
    """MeanAggregation class for calculating the mean of values in a specified column.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Aggregate data by calculating the mean.
    """

    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Aggregate data by calculating the mean.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        return df.groupby(*key_cols).agg(F.mean(F.col(agg_col)))

aggregate(df, agg_col, key_cols)

Aggregate data by calculating the mean.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Aggregate data by calculating the mean.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    return df.groupby(*key_cols).agg(F.mean(F.col(agg_col)))

StddevAggregation

Bases: AggregationStrategy

StddevAggregation class for calculating the standard deviation of values in a specified column.

Methods:

Name Description
aggregate

Aggregate data by calculating the standard deviation.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
class StddevAggregation(AggregationStrategy):
    """StddevAggregation class for calculating the standard deviation of values in a specified column.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Aggregate data by calculating the standard deviation.
    """

    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Aggregate data by calculating the standard deviation.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        return df.groupby(*key_cols).agg(F.stddev(F.col(agg_col)))

aggregate(df, agg_col, key_cols)

Aggregate data by calculating the standard deviation.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Aggregate data by calculating the standard deviation.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    return df.groupby(*key_cols).agg(F.stddev(F.col(agg_col)))

SumAggregation

Bases: AggregationStrategy

SumAggregation class for summing values in a specified column.

Methods:

Name Description
aggregate

Aggregate data by summing the values.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
class SumAggregation(AggregationStrategy):
    """SumAggregation class for summing values in a specified column.

    Methods
    -------
    aggregate(df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame
        Aggregate data by summing the values.
    """

    def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
        """Aggregate data by summing the values.

        Parameters
        ----------
        df : DataFrame
            The input DataFrame containing the data to be aggregated.
        agg_col : str
            The column in the DataFrame to be aggregated.
        key_cols : List[str]
            The columns in the DataFrame to group by.

        Returns
        -------
        DataFrame
            A DataFrame containing the aggregated data.
        """
        return df.groupby(*key_cols).agg(F.sum(F.col(agg_col)))

aggregate(df, agg_col, key_cols)

Aggregate data by summing the values.

Parameters:

Name Type Description Default
df DataFrame

The input DataFrame containing the data to be aggregated.

 required
agg_col str

The column in the DataFrame to be aggregated.

 required
key_cols List[str]

The columns in the DataFrame to group by.

 required

Returns:

Type Description
DataFrame

A DataFrame containing the aggregated data.
Source code in amee_utils/feature_generator/feature_set/aggregation_strategy.py 
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
def aggregate(self, df: DataFrame, agg_col: str, key_cols: List[str]) -> DataFrame:
    """Aggregate data by summing the values.

    Parameters
    ----------
    df : DataFrame
        The input DataFrame containing the data to be aggregated.
    agg_col : str
        The column in the DataFrame to be aggregated.
    key_cols : List[str]
        The columns in the DataFrame to group by.

    Returns
    -------
    DataFrame
        A DataFrame containing the aggregated data.
    """
    return df.groupby(*key_cols).agg(F.sum(F.col(agg_col)))