ModelingModeConfig#
- class datarobotx.ModelingModeConfig(metric=None, mode=None, use_time_series=None, unsupervised_mode=None, unsupervised_type=None, shap_only_mode=None, quantile_level=None, accuracy_optimized_mb=None, only_include_monotonic_blueprints=None, monotonic_increasing_featurelist_id=None, monotonic_decreasing_featurelist_id=None)#
Autopilot modes and metric(s).
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:
metric (str) – The metric to use to select the best models. See /api/v2/projects/(projectId)/features/metrics/ for the metrics that may be valid for a potential target. Note that weighted metrics must be used with a weights column.
mode ({'4', 'manual', '2', '0', 'auto', 'comprehensive', '3', 'quick'}) – The autopilot mode to use. Either ‘quick’, ‘auto’, ‘manual’ or ‘comprehensive’
use_time_series (bool) – A boolean value indicating whether a time series project should be created instead of a regular project which uses datetime partitioning.
unsupervised_mode (bool) – If True, unsupervised project (without target) will be created.
target
cannot be specified ifunsupervisedMode
is True.unsupervised_type ({'anomaly', 'clustering'}) – The type of unsupervised project. Only valid when unsupervisedMode is true. If unsupervisedMode, defaults to anomaly.
shap_only_mode (bool) – Keep only models that support SHAP values during Autopilot run. Use SHAP- based insights wherever possible.
quantile_level (float) – The quantile level between 0.01 and 0.99 for specifying the Quantile metric.
accuracy_optimized_mb (bool) – Include additional, longer-running models that will be run by the autopilot and available to run manually.
only_include_monotonic_blueprints (bool) – When true, only blueprints that support enforcing montonic constraints will be available in the project or selected for autopilot.
monotonic_increasing_featurelist_id (str) – The ID of the featurelist that defines the set of features with a monotonically increasing relationship to the target. If null, no such constraints are enforced. When specified, this will set a default for the project that can be overridden at model submission time if desired.
monotonic_decreasing_featurelist_id (str) – The ID of the featurelist that defines the set of features with a monotonically decreasing relationship to the target. If null, no such constraints are enforced. When specified, this will set a default for the project that can be overridden at model submission time if desired.
See also
DRConfig
Configuration object for DataRobot project and autopilot settings, also includes detailed examples of usage
Attributes:
Include additional, longer-running models that will be run by the autopilot and available to run manually.
The metric to use to select the best models.
The autopilot mode to use.
The ID of the featurelist that defines the set of features with a monotonically decreasing relationship to the target.
The ID of the featurelist that defines the set of features with a monotonically increasing relationship to the target.
When true, only blueprints that support enforcing montonic constraints will be available in the project or selected for autopilot.
The quantile level between 0.01 and 0.99 for specifying the Quantile metric.
Keep only models that support SHAP values during Autopilot run.
If True, unsupervised project (without target) will be created.
The type of unsupervised project.
A boolean value indicating whether a time series project should be created instead of a regular project which uses datetime partitioning.
Inherited methods:
keys
()- rtype:
to_dict
()Return configuration as a dict.
- property accuracy_optimized_mb: bool#
Include additional, longer-running models that will be run by the autopilot and available to run manually.
Notes
accuracy_optimized_mb : bool
- property metric: str#
The metric to use to select the best models. See /api/v2/projects/(projectId)/features/metrics/ for the metrics that may be valid for a potential target. Note that weighted metrics must be used with a weights column.
Notes
metric : str
- property mode: str#
The autopilot mode to use. Either ‘quick’, ‘auto’, ‘manual’ or ‘comprehensive’.
Notes
mode : {‘4’, ‘manual’, ‘2’, ‘0’, ‘auto’, ‘comprehensive’, ‘3’, ‘quick’}
- property monotonic_decreasing_featurelist_id: str#
The ID of the featurelist that defines the set of features with a monotonically decreasing relationship to the target. If null, no such constraints are enforced. When specified, this will set a default for the project that can be overridden at model submission time if desired.
Notes
monotonic_decreasing_featurelist_id : str
- property monotonic_increasing_featurelist_id: str#
The ID of the featurelist that defines the set of features with a monotonically increasing relationship to the target. If null, no such constraints are enforced. When specified, this will set a default for the project that can be overridden at model submission time if desired.
Notes
monotonic_increasing_featurelist_id : str
- property only_include_monotonic_blueprints: bool#
When true, only blueprints that support enforcing montonic constraints will be available in the project or selected for autopilot.
Notes
only_include_monotonic_blueprints : bool
- property quantile_level: float#
The quantile level between 0.01 and 0.99 for specifying the Quantile metric.
Notes
quantile_level : float
- property shap_only_mode: bool#
Keep only models that support SHAP values during Autopilot run. Use SHAP-based insights wherever possible.
Notes
shap_only_mode : bool
- property unsupervised_mode: bool#
If True, unsupervised project (without target) will be created.
target
cannot be specified ifunsupervisedMode
is True.Notes
unsupervised_mode : bool