getml.feature_learning.aggregations
COUNT_DISTINCT module-attribute
COUNT_DISTINCT: Final[CountDistinct] = 'COUNT DISTINCT'
Count function with distinct clause. This only counts unique elements.
COUNT_DISTINCT_OVER_COUNT module-attribute
COUNT_DISTINCT_OVER_COUNT: Final[CountDistinctOverCount] = (
"COUNT DISTINCT OVER COUNT"
)
COUNT DISTINCT divided by COUNT. Please note that this aggregation is not supported by Multirel
.
COUNT_MINUS_COUNT_DISTINCT module-attribute
COUNT_MINUS_COUNT_DISTINCT: Final[
CountMinusCountDistinct
] = "COUNT MINUS COUNT DISTINCT"
Counts minus counts distinct. Substracts COUNT DISTINCT from COUNT.
EWMA_1D module-attribute
Exponentially weighted moving average with a half-life of 1 day. Please note that this aggregation is not supported by Multirel
.
EWMA_1H module-attribute
Exponentially weighted moving average with a half-life of 1 hour. Please note that this aggregation is not supported by Multirel
.
EWMA_1M module-attribute
Exponentially weighted moving average with a half-life of 1 minute. Please note that this aggregation is not supported by Multirel
.
EWMA_1S module-attribute
Exponentially weighted moving average with a half-life of 1 second.
EWMA_7D module-attribute
Exponentially weighted moving average with a half-life of 7 days. Please note that this aggregation is not supported by Multirel
.
EWMA_30D module-attribute
Exponentially weighted moving average with a half-life of 30 days. Please note that this aggregation is not supported by Multirel
.
EWMA_90D module-attribute
Exponentially weighted moving average with a half-life of 90 days. Please note that this aggregation is not supported by Multirel
.
EWMA_365D module-attribute
Exponentially weighted moving average with a half-life of 365 days. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_1D module-attribute
EWMA_TREND_1D: Final[EWMA_TREND_1d] = 'EWMA_TREND_1D'
Exponentially weighted trend with a half-life of 1 day. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_1H module-attribute
EWMA_TREND_1H: Final[EWMA_TREND_1h] = 'EWMA_TREND_1H'
Exponentially weighted trend with a half-life of 1 hour. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_1M module-attribute
EWMA_TREND_1M: Final[EWMA_TREND_1m] = 'EWMA_TREND_1M'
Exponentially weighted trend with a half-life of 1 minute. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_1S module-attribute
EWMA_TREND_1S: Final[EWMA_TREND_1s] = 'EWMA_TREND_1S'
Exponentially weighted trend with a half-life of 1 second.
EWMA_TREND_7D module-attribute
EWMA_TREND_7D: Final[EWMA_TREND_7d] = 'EWMA_TREND_7D'
Exponentially weighted trend with a half-life of 7 days. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_30D module-attribute
EWMA_TREND_30D: Final[EWMA_TREND_30d] = 'EWMA_TREND_30D'
Exponentially weighted trend with a half-life of 30 days. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_90D module-attribute
EWMA_TREND_90D: Final[EWMA_TREND_90d] = 'EWMA_TREND_90D'
Exponentially weighted trend with a half-life of 90 days. Please note that this aggregation is not supported by Multirel
.
EWMA_TREND_365D module-attribute
EWMA_TREND_365D: Final[EWMA_TREND_365d] = 'EWMA_TREND_365D'
Exponentially weighted trend with a half-life of 365 days. Please note that this aggregation is not supported by Multirel
.
FIRST module-attribute
First value of a given column, when ordered by the time stamp.
KURTOSIS module-attribute
The kurtosis of a given column. Please note that this aggregation is not supported by Multirel
.
LAST module-attribute
Last value of a given column, when ordered by the time stamp.
MODE module-attribute
Most frequent value of a given column. Please note that this aggregation is not supported by Multirel
.
NUM_MAX module-attribute
The number of times we observe the maximum value. Please note that this aggregation is not supported by Multirel
.
NUM_MIN module-attribute
The number of times we observe the minimum value. Please note that this aggregation is not supported by Multirel
.
Q_1 module-attribute
The 1%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_5 module-attribute
The 5%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_10 module-attribute
The 10%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_25 module-attribute
The 25%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_75 module-attribute
The 75%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_90 module-attribute
The 90%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_95 module-attribute
The 95%-quantile. Please note that this aggregation is not supported by Multirel
.
Q_99 module-attribute
The 99%-quantile. Please note that this aggregation is not supported by Multirel
.
SKEW module-attribute
Skewness of a given column. Please note that this aggregation is not supported by Multirel
.
TIME_SINCE_FIRST_MAXIMUM module-attribute
TIME_SINCE_FIRST_MAXIMUM: Final[TimeSinceFirstMaximum] = (
"TIME SINCE FIRST MAXIMUM"
)
The time difference between the first time we see the maximum value and the time stamp in the population table. Please note that this aggregation is not supported by Multirel
.
TIME_SINCE_FIRST_MINIMUM module-attribute
TIME_SINCE_FIRST_MINIMUM: Final[TimeSinceFirstMinimum] = (
"TIME SINCE FIRST MINIMUM"
)
The time difference between the first time we see the minimum value and the time stamp in the population table. Please note that this aggregation is not supported by Multirel
.
TIME_SINCE_LAST_MAXIMUM module-attribute
TIME_SINCE_LAST_MAXIMUM: Final[TimeSinceLastMaximum] = (
"TIME SINCE LAST MAXIMUM"
)
The time difference between the last time we see the maximum value and the time stamp in the population table. Please note that this aggregation is not supported by Multirel
.
TIME_SINCE_LAST_MINIMUM module-attribute
TIME_SINCE_LAST_MINIMUM: Final[TimeSinceLastMinimum] = (
"TIME SINCE LAST MINIMUM"
)
The time difference between the last time we see the minimum value and the time stamp in the population table. Please note that this aggregation is not supported by Multirel
.
TREND module-attribute
Extracts a linear trend from a variable over time and extrapolates this trend to the current time stamp. Please note that this aggregation is not supported by Multirel
.
VAR module-attribute
Statistical variance of a given numerical column. Please note that this aggregation is not supported by Multirel
.
VARIATION_COEFFICIENT module-attribute
VARIATION_COEFFICIENT: Final[VariationCoefficient] = (
"VARIATION COEFFICIENT"
)
VAR divided by MEAN. Please note that this aggregation is not supported by Multirel
.
AGGREGATIONS module-attribute
AGGREGATIONS: FrozenSet[Aggregations] = (
MULTIREL_AGGREGATIONS
| FASTPROP_AGGREGATIONS
| MAPPING_AGGREGATIONS
)
Set of all possible aggregations.
FASTPROP module-attribute
FASTPROP = FastPropAggregationsSets(
all=FASTPROP_AGGREGATIONS,
default=frozenset(
{
AVG,
COUNT,
COUNT_DISTINCT,
COUNT_MINUS_COUNT_DISTINCT,
FIRST,
LAST,
MAX,
MEDIAN,
MIN,
MODE,
STDDEV,
SUM,
TREND,
}
),
minimal=frozenset([AVG, COUNT, MAX, MIN, SUM]),
)
Set of default aggregations for FastProp
. all
contains all aggregations supported by FastProp, default
contains the subset of reasonable default aggregations, minimal
is minimal set.
FASTPROP_AGGREGATIONS module-attribute
FASTPROP_AGGREGATIONS: FrozenSet[FastPropAggregations] = (
frozenset(
{
AVG,
COUNT,
COUNT_DISTINCT,
COUNT_MINUS_COUNT_DISTINCT,
FIRST,
LAST,
MAX,
MEDIAN,
MIN,
STDDEV,
SUM,
VAR,
COUNT_DISTINCT_OVER_COUNT,
EWMA_1S,
EWMA_1M,
EWMA_1H,
EWMA_1D,
EWMA_7D,
EWMA_30D,
EWMA_90D,
EWMA_365D,
EWMA_TREND_1S,
EWMA_TREND_1M,
EWMA_TREND_1H,
EWMA_TREND_1D,
EWMA_TREND_7D,
EWMA_TREND_30D,
EWMA_TREND_90D,
EWMA_TREND_365D,
KURTOSIS,
MODE,
NUM_MAX,
NUM_MIN,
Q_1,
Q_5,
Q_10,
Q_25,
Q_75,
Q_90,
Q_95,
Q_99,
SKEW,
TIME_SINCE_FIRST_MAXIMUM,
TIME_SINCE_FIRST_MINIMUM,
TIME_SINCE_LAST_MAXIMUM,
TIME_SINCE_LAST_MINIMUM,
TREND,
VARIATION_COEFFICIENT,
}
)
)
Set of all aggregations supported by FastProp.
MAPPING module-attribute
MAPPING = MappingAggregationsSets(
all=MAPPING_AGGREGATIONS,
default=frozenset({AVG}),
minimal=frozenset({AVG}),
)
Set of default aggregations for Mapping
. all
contains all aggregations supported by the mapping preprocessor. default
and minimal
are identical and include only the AVG aggregation, which is the recommended setting for classification problems.
MAPPING_AGGREGATIONS module-attribute
MAPPING_AGGREGATIONS: FrozenSet[MappingAggregations] = (
frozenset(
{
AVG,
COUNT,
COUNT_DISTINCT,
COUNT_DISTINCT_OVER_COUNT,
COUNT_MINUS_COUNT_DISTINCT,
KURTOSIS,
MAX,
MEDIAN,
MIN,
MODE,
NUM_MAX,
NUM_MIN,
Q_1,
Q_5,
Q_10,
Q_25,
Q_75,
Q_90,
Q_95,
Q_99,
SKEW,
STDDEV,
SUM,
VAR,
VARIATION_COEFFICIENT,
}
)
)
Set of all aggregations supported by the mapping preprocessor.
MULTIREL module-attribute
MULTIREL = MultirelAggregationsSets(
all=MULTIREL_AGGREGATIONS,
default=frozenset({AVG, COUNT, MAX, MIN, SUM}),
minimal=frozenset([AVG, COUNT, SUM]),
)
Set of default aggregations for Multirel
. all
contains all aggregations supported by Multirel, default
contains the subset of reasonable default aggregations, minimal
is minimal set.
MULTIREL_AGGREGATIONS module-attribute
MULTIREL_AGGREGATIONS: FrozenSet[MultirelAggregations] = (
frozenset(
{
AVG,
COUNT,
COUNT_DISTINCT,
COUNT_MINUS_COUNT_DISTINCT,
FIRST,
LAST,
MAX,
MEDIAN,
MIN,
STDDEV,
SUM,
VAR,
}
)
)
Set of all aggregations supported by Multirel.