PartitioningDateTimeConfig

class datarobotx.PartitioningDateTimeConfig(datetime_partition_column=None, validation_duration=None, gap_duration=None, is_holdout_modified=None, holdout_start_date=None, holdout_end_date=None, holdout_duration=None, autopilot_data_selection_method=None, autopilot_data_sampling_method=None, number_of_backtests=None, use_project_settings=None, backtests=None)

Date-time partitioning configuration.

Parameters that default to ‘None’ (or are omitted by the user) are overridden to server-side defaults at runtime. Consult the DataRobot REST API and GUI documentation for additional information on each parameter.

Parameters:

• datetime_partition_column (str) – The date column that will be used as a datetime partition column.

• validation_duration (str) – The default validation duration for all backtests. If the primary date/time feature in a time series project is irregular, you cannot set a default validation length. Instead, set each duration individually. For an OTV project setting the validation duration will always use regular partitioning. Omitting it will use irregular partitioning if the date/time feature is irregular.

• gap_duration (str) – The duration of the gap between holdout training and holdout scoring data. For time series projects, defaults to the duration of the gap between the end of the feature derivation window and the beginning of the forecast window. For OTV projects, defaults to a zero duration (P0Y0M0D).

• is_holdout_modified (bool) – A boolean value indicating whether holdout settings (start/end dates) have been modified by user.

• holdout_start_date (str) – The start date of holdout scoring data. When specifying holdoutStartDate, one of holdoutEndDate or holdoutDuration must also be specified. This attribute cannot be specified when disableHoldout is true.

• holdout_end_date (str) – The end date of holdout scoring data. When specifying holdoutEndDate, holdoutStartDate must also be specified. This attribute cannot be specified when disableHoldout is true.

• holdout_duration (str) – The duration of holdout scoring data. When specifying holdoutDuration, holdoutStartDate must also be specified. This attribute cannot be specified when disableHoldout is true.

• autopilot_data_selection_method ({'duration', 'rowCount'}) – The Data Selection method to be used by autopilot when creating models for datetime-partitioned datasets.

• autopilot_data_sampling_method ({'random', 'latest'}) – Defines how autopilot will select subsample from training dataset in OTV/TS projects. Defaults to ‘latest’ for ‘rowCount’ dataSelectionMethod and to ‘random’ for ‘duration’.

• number_of_backtests (int) – The number of backtests to use. If omitted, defaults to a positive value selected by the server based on the validation and gap durations.

• use_project_settings (bool) – Specifies whether datetime-partitioned project should use project settings (i.e. backtests configuration has been modified by the user).

• backtests (list of PartitioningDTBacktestConfig) – An array specifying the format of the backtests.

See also

DRConfig

Configuration object for DataRobot project and autopilot settings, also includes detailed examples of usage

Attributes:

autopilot_data_sampling_method	Defines how autopilot will select subsample from training dataset in OTV/TS projects.
autopilot_data_selection_method	The Data Selection method to be used by autopilot when creating models for datetime-partitioned datasets.
backtests	An array specifying the format of the backtests.
datetime_partition_column	The date column that will be used as a datetime partition column.
gap_duration	The duration of the gap between holdout training and holdout scoring data.
holdout_duration	The duration of holdout scoring data.
holdout_end_date	The end date of holdout scoring data.
holdout_start_date	The start date of holdout scoring data.
is_holdout_modified	A boolean value indicating whether holdout settings (start/end dates) have been modified by user.
number_of_backtests	The number of backtests to use.
use_project_settings	Specifies whether datetime-partitioned project should use project settings (i.e.
validation_duration	The default validation duration for all backtests.

Inherited methods:

keys()	rtype: Collection[str]
to_dict()	Return configuration as a dict.

property autopilot_data_sampling_method: str

Defines how autopilot will select subsample from training dataset in OTV/TS projects. Defaults to ‘latest’ for ‘rowCount’ dataSelectionMethod and to ‘random’ for ‘duration’.

Notes

autopilot_data_sampling_method : {‘random’, ‘latest’}

property autopilot_data_selection_method: str

The Data Selection method to be used by autopilot when creating models for datetime-partitioned datasets.

Notes

autopilot_data_selection_method : {‘duration’, ‘rowCount’}

property backtests: List[PartitioningDTBacktestConfig]

An array specifying the format of the backtests.

Notes

backtests : list of PartitioningDTBacktestConfig

property datetime_partition_column: str

The date column that will be used as a datetime partition column.

Notes

datetime_partition_column : str

property gap_duration: str

The duration of the gap between holdout training and holdout scoring data. For time series projects, defaults to the duration of the gap between the end of the feature derivation window and the beginning of the forecast window. For OTV projects, defaults to a zero duration (P0Y0M0D).

Notes

gap_duration : str

property holdout_duration: str

The duration of holdout scoring data. When specifying holdoutDuration, holdoutStartDate must also be specified. This attribute cannot be specified when disableHoldout is true.

Notes

holdout_duration : str

property holdout_end_date: str

The end date of holdout scoring data. When specifying holdoutEndDate, holdoutStartDate must also be specified. This attribute cannot be specified when disableHoldout is true.

Notes

holdout_end_date : str

property holdout_start_date: str

The start date of holdout scoring data. When specifying holdoutStartDate, one of holdoutEndDate or holdoutDuration must also be specified. This attribute cannot be specified when disableHoldout is true.

Notes

holdout_start_date : str

property is_holdout_modified: bool

A boolean value indicating whether holdout settings (start/end dates) have been modified by user.

Notes

is_holdout_modified : bool

property number_of_backtests: int

The number of backtests to use. If omitted, defaults to a positive value selected by the server based on the validation and gap durations.

Notes

number_of_backtests : int

to_dict()

Return configuration as a dict.

Return type:

Dict[str, Any]

property use_project_settings: bool

Specifies whether datetime-partitioned project should use project settings (i.e. backtests configuration has been modified by the user).

Notes

use_project_settings : bool

property validation_duration: str

The default validation duration for all backtests. If the primary date/time feature in a time series project is irregular, you cannot set a default validation length. Instead, set each duration individually. For an OTV project setting the validation duration will always use regular partitioning. Omitting it will use irregular partitioning if the date/time feature is irregular.

Notes

validation_duration : str