FeatureAttributes#
Overview#
The mapping of attributes for a single feature
- type:
object
Properties#
auto_derive_on_train#
Derive feature by creating two new continuous features: .series_progress and .series_progress_delta. Series progress values range from 0 to 1.0 for each case in the series. Series progress delta values are the delta value of the progress for each case. Both of these features are used to determine when to stop series synthesis.
- type:
object
Nested Properties:#
code#
The Amalgam code used to derive the feature value.
- type:
string
derive_type#
The train derive operation type.
- type:
string
- enum:
custom | start | end | progress
ordered_by_features#
Feature name(s) that define the order of the series.
- type:
list
- values:
string
series_id_features#
Feature name(s) whose values are used to identify cases within the same series.
- type:
list
- values:
string
bounds#
- ref:
cycle_length#
Cyclic features are set by specifying a cycle_length value in the feature attributes. cycle_length requires a single value, which is the upper bound of the difference for the cycle range. For example, if cycle_length is 360, then a value of 1 and 359 will have a difference of 2. Cyclic features have no restrictions in the input dataset, however, cyclic features will be output on a scale from 0 to cycle_length. To constrain the output to a different range, modify the min and max bounds feature attribute.
Examples: - degrees: values should be 0-359, cycle_length = 360 - days: values should be 0-6, cycle_length = 7 - hours: values should be 0-23, cycle_length = 24
- type:
number
data_type#
- ref:
date_time_format#
If specified, feature values should match the date format specified by this string. Only applicable to continuous features.
- type:
string
decimal_places#
Decimal places to round to, default is no rounding. If significant_digits is also specified, the number will be rounded to the specified number of significant digits first, then rounded to the number of decimal points as specified by this parameter.
- type:
number
dependent_features#
A list of other feature names that this feature either depends on or features that depend on this feature. Should be used when there are multi-type value features that tightly depend on values based on other multi-type value features.
- type:
list
- values:
string
derived_feature_code#
Code defining how the value for this feature could be derived if this feature is specified as a derived_context_feature or a derived_action_feature during react flows. For react_series, the data referenced is the accumulated series data (as a list of rows), and for non-series reacts, the data is the one single row. Each row is comprised of all the combined context and action features. Referencing data in these rows uses 0-based indexing, where the current row index is 0, the previous row’s is 1, etc. The specified code may do simple logic and numeric operations on feature values referenced via feature name and row offset
Examples:
- "#x 1": Use the value for feature ‘x’ from the previously processed row (offset of 1, one lag value).
- "(- #y 0 #x 1)": Feature ‘y’ value from current (offset 0) row minus feature ‘x’ value from previous (offset 1) row.
- type:
string
id_feature#
Set to true for nominal features containing nominal IDs, specifying that his feature should be used to compute case weights for id based privacy. For time series, this feature will be used as the id for each time series generation.
- type:
boolean
locale#
The date time format locale. If unspecified, uses platform default locale.
- type:
string
max_row_lag#
The number of time steps traced back by the maximum lag feature created for this feature.
- type:
number
non_sensitive#
Flag a categorical nominal feature as non-sensitive. It is recommended that all nominal features be represented with either an int-id subtype or another available nominal subtype using the subtype attribute. However, if the nominal feature is non-sensitive, setting this parameter to true will bypass the subtype requirement. Only applicable to nominal features.
- type:
boolean
null_is_dependent#
Modify how dependent features with nulls are treated during a react, specifically when they use null as a context value. Only applicable to dependent features.
When false (default), the feature will be treated as a non-dependent context feature. When true for nominal types, treats null as an individual dependent class value, only cases that also have nulls as this feature’s value will be considered. When true for continuous types, only the cases with the same dependent feature values as the cases that also have nulls as this feature’s value will be considered.
- type:
boolean
observational_error#
Specifies the observational mean absolute error for this feature. Use when the error value is already known.
- type:
number
original_format#
Original data formats used by clients. Automatically populated by clients to store client language specific context about features.
- type:
any
original_type#
Original data type details. Used by clients to determine how to serialize and deserialize feature data.
- ref:
parent#
The feature whose values this time-series feature’s values are derived from.
- type:
string
parent_type#
The type of time-series processing used by the parent feature.
- type:
string
- enum:
delta | rate
post_process#
Custom Amalgam code that is called on resulting values of this feature during react operations.
- type:
string
sample#
A sample of a value for the feature.
- type:
any
significant_digits#
Round to the specified significant digits, default is no rounding.
- type:
number
subtype#
The type used in novel nominal substitution.
- type:
string
time_series#
Time series options for a feature.
- type:
object
Nested Properties:#
delta_max#
If specified, ensures that the largest difference between feature values is not larger than this specified value. A null value means no max boundary. The length of the list must match the number of derivatives as specified by order. Only applicable when time series type is set to delta.
- type:
list
- values:
number
delta_min#
If specified, ensures that the smallest difference between features values is not smaller than this specified value. A null value means no min boundary. The length of the list must match the number of derivatives as specified by order. Only applicable when time series type is set to delta.
- type:
list
- values:
number
derived_orders#
The number of orders of derivatives that should be derived instead of synthesized. Ignored if order is not provided.
- type:
number
lags#
If specified, generates lag features containing previous values using the enumerated lag offsets. Takes precedence over num_lags. If neither num_lags nor lags is specified for a feature, then a single lag feature is generated.
- type:
list
- values:
number
num_lags#
If specified, generates the specified amount of lag features containing previous values. If lags is specified, then this parameter will be ignored. If neither num_lags nor lags is specified for a feature, then a single lag feature is generated.
- type:
number
order#
If provided, will generate the specified number of derivatives and boundary values.
- type:
number
rate_max#
If specified, ensures that the rate (the difference quotient, the discrete version of derivative) for this feature won’t be more than the value provided. A null value means no max boundary. The value must be in epoch format for the time feature. The length of the list must match the number of derivatives as specified by order. Only applicable when time series type is set to rate.
- type:
list
- values:
number
rate_min#
If specified, ensures that the rate (the difference quotient, the discrete version of derivative) for this feature won’t be less than the value provided. A null value means no min boundary. The value must be in epoch format for the time feature. The length of the list must match the number of derivatives as specified by order. Only applicable when time series type is set to rate.
- type:
list
- values:
number
series_has_terminators#
When true, requires that the model identify and learn values that explicitly denote the end of a series. Only applicable to id features for a series.
- type:
boolean
stop_on_terminator#
When true, requires that a series ends on a terminator value. Only applicable to id features for a series.
- type:
boolean
time_feature#
When true, the feature will be treated as the time feature for time series modeling. Additionally, time features must use type delta.
- type:
boolean
type#
When rate is specified, uses the difference of the current value from its previous value divided by the change in time since the previous value. When delta is specified, uses the difference of the current value from its previous value regardless of the elapsed time. Set to delta if feature has time_feature set to true.
- type:
string
- enum:
rate | delta
universal#
Controls whether future values of independent time series are considered. Applicable only to the time feature. When false, the time feature is not universal and allows using future data from other series in decisions; this is applicable when the time is not globally relevant and is independent for each time series. When true, universally excludes using any data with from the future from all series; this is applicable when time is globally relevant and there are events that may affect all time series. If there is any possibility of global relevancy of time, it is generally recommended to set this value to true, which is the default.
- type:
boolean
ts_order#
The order of rate/delta being described by this time-series feature.
- type:
number
- min:
0
ts_type#
The type of value being captured by this time-series feature.
- type:
string
- enum:
lag | delta | rate
type#
- ref:
- required:
True
unique#
Flag feature as only having unique values. Only applicable to nominals features.
- type:
boolean