samplers¶
Submodules¶
Package Contents¶
Classes Summary¶
SMOTENC Oversampler component. Uses SMOTENC to generate synthetic samples. Works on a mix of nomerical and categorical columns. |
|
SMOTEN Oversampler component. Uses SMOTEN to generate synthetic samples. Works for purely categorical datasets. |
|
SMOTE Oversampler component. Works on numerical datasets only. This component is only run during training and not during predict. |
|
Initializes an undersampling transformer to downsample the majority classes in the dataset. |
Contents¶
-
class
evalml.pipelines.components.transformers.samplers.SMOTENCSampler(sampling_ratio=0.25, k_neighbors_default=5, n_jobs=- 1, random_seed=0, **kwargs)[source]¶ SMOTENC Oversampler component. Uses SMOTENC to generate synthetic samples. Works on a mix of nomerical and categorical columns. Input data must be Woodwork type, and this component is only run during training and not during predict.
- Parameters
sampling_ratio (float) – This is the goal ratio of the minority to majority class, with range (0, 1]. A value of 0.25 means we want a 1:4 ratio of the minority to majority class after oversampling. We will create the a sampling dictionary using this ratio, with the keys corresponding to the class and the values responding to the number of samples. Defaults to 0.25.
k_neighbors_default (int) – The number of nearest neighbors used to construct synthetic samples. This is the default value used, but the actual k_neighbors value might be smaller if there are less samples. Defaults to 5.
n_jobs (int) – The number of CPU cores to use. Defaults to -1.
random_seed (int) – The seed to use for random sampling. Defaults to 0.
Attributes
hyperparameter_ranges
{}
model_family
ModelFamily.NONE
modifies_features
True
modifies_target
True
name
SMOTENC Oversampler
Methods
Constructs a new component with the same parameters and random state.
Returns the default parameters for this component.
Describe a component and its parameters
Fits the Oversampler to the data.
Fit and transform the data using the data sampler. Used during training of the pipeline
Loads component at file path
Returns boolean determining if component needs fitting before
Returns the parameters which were used to initialize the component
Saves component at file path
No transformation needs to be done here.
-
clone(self)¶ Constructs a new component with the same parameters and random state.
- Returns
A new instance of this component with identical parameters and random state.
-
default_parameters(cls)¶ Returns the default parameters for this component.
Our convention is that Component.default_parameters == Component().parameters.
- Returns
default parameters for this component.
- Return type
dict
-
describe(self, print_name=False, return_dict=False)¶ Describe a component and its parameters
- Parameters
print_name (bool, optional) – whether to print name of component
return_dict (bool, optional) – whether to return description as dictionary in the format {“name”: name, “parameters”: parameters}
- Returns
prints and returns dictionary
- Return type
None or dict
-
fit(self, X, y)[source]¶ Fits the Oversampler to the data.
- Parameters
X (pd.DataFrame) – Training features
y (pd.Series) – Target features
- Returns
self
-
fit_transform(self, X, y)¶ Fit and transform the data using the data sampler. Used during training of the pipeline
- Parameters
X (pd.DataFrame) – Training features
y – Target features
-
static
load(file_path)¶ Loads component at file path
- Parameters
file_path (str) – Location to load file
- Returns
ComponentBase object
-
needs_fitting(self)¶ Returns boolean determining if component needs fitting before calling predict, predict_proba, transform, or feature_importances. This can be overridden to False for components that do not need to be fit or whose fit methods do nothing.
-
property
parameters(self)¶ Returns the parameters which were used to initialize the component
-
save(self, file_path, pickle_protocol=cloudpickle.DEFAULT_PROTOCOL)¶ Saves component at file path
- Parameters
file_path (str) – Location to save file
pickle_protocol (int) – The pickle data stream format.
- Returns
None
-
transform(self, X, y=None)¶ No transformation needs to be done here.
- Parameters
X (pd.DataFrame) – Training features. Ignored.
y (pd.Series) – Target features. Ignored.
- Returns
X and y data that was passed in.
- Return type
pd.DataFrame, pd.Series
-
class
evalml.pipelines.components.transformers.samplers.SMOTENSampler(sampling_ratio=0.25, k_neighbors_default=5, n_jobs=- 1, random_seed=0, **kwargs)[source]¶ SMOTEN Oversampler component. Uses SMOTEN to generate synthetic samples. Works for purely categorical datasets. This component is only run during training and not during predict.
- Parameters
sampling_ratio (float) – This is the goal ratio of the minority to majority class, with range (0, 1]. A value of 0.25 means we want a 1:4 ratio of the minority to majority class after oversampling. We will create the a sampling dictionary using this ratio, with the keys corresponding to the class and the values responding to the number of samples. Defaults to 0.25.
k_neighbors_default (int) – The number of nearest neighbors used to construct synthetic samples. This is the default value used, but the actual k_neighbors value might be smaller if there are less samples. Defaults to 5.
n_jobs (int) – The number of CPU cores to use. Defaults to -1.
random_seed (int) – The seed to use for random sampling. Defaults to 0.
Attributes
hyperparameter_ranges
{}
model_family
ModelFamily.NONE
modifies_features
True
modifies_target
True
name
SMOTEN Oversampler
Methods
Constructs a new component with the same parameters and random state.
Returns the default parameters for this component.
Describe a component and its parameters
Fits the Oversampler to the data.
Fit and transform the data using the data sampler. Used during training of the pipeline
Loads component at file path
Returns boolean determining if component needs fitting before
Returns the parameters which were used to initialize the component
Saves component at file path
No transformation needs to be done here.
-
clone(self)¶ Constructs a new component with the same parameters and random state.
- Returns
A new instance of this component with identical parameters and random state.
-
default_parameters(cls)¶ Returns the default parameters for this component.
Our convention is that Component.default_parameters == Component().parameters.
- Returns
default parameters for this component.
- Return type
dict
-
describe(self, print_name=False, return_dict=False)¶ Describe a component and its parameters
- Parameters
print_name (bool, optional) – whether to print name of component
return_dict (bool, optional) – whether to return description as dictionary in the format {“name”: name, “parameters”: parameters}
- Returns
prints and returns dictionary
- Return type
None or dict
-
fit(self, X, y)¶ Fits the Oversampler to the data.
- Parameters
X (pd.DataFrame) – Training features
y (pd.Series) – Target features
- Returns
self
-
fit_transform(self, X, y)¶ Fit and transform the data using the data sampler. Used during training of the pipeline
- Parameters
X (pd.DataFrame) – Training features
y – Target features
-
static
load(file_path)¶ Loads component at file path
- Parameters
file_path (str) – Location to load file
- Returns
ComponentBase object
-
needs_fitting(self)¶ Returns boolean determining if component needs fitting before calling predict, predict_proba, transform, or feature_importances. This can be overridden to False for components that do not need to be fit or whose fit methods do nothing.
-
property
parameters(self)¶ Returns the parameters which were used to initialize the component
-
save(self, file_path, pickle_protocol=cloudpickle.DEFAULT_PROTOCOL)¶ Saves component at file path
- Parameters
file_path (str) – Location to save file
pickle_protocol (int) – The pickle data stream format.
- Returns
None
-
transform(self, X, y=None)¶ No transformation needs to be done here.
- Parameters
X (pd.DataFrame) – Training features. Ignored.
y (pd.Series) – Target features. Ignored.
- Returns
X and y data that was passed in.
- Return type
pd.DataFrame, pd.Series
-
class
evalml.pipelines.components.transformers.samplers.SMOTESampler(sampling_ratio=0.25, k_neighbors_default=5, n_jobs=- 1, random_seed=0, **kwargs)[source]¶ SMOTE Oversampler component. Works on numerical datasets only. This component is only run during training and not during predict.
- Parameters
sampling_ratio (float) – This is the goal ratio of the minority to majority class, with range (0, 1]. A value of 0.25 means we want a 1:4 ratio of the minority to majority class after oversampling. We will create the a sampling dictionary using this ratio, with the keys corresponding to the class and the values responding to the number of samples. Defaults to 0.25.
k_neighbors_default (int) – The number of nearest neighbors used to construct synthetic samples. This is the default value used, but the actual k_neighbors value might be smaller if there are less samples. Defaults to 5.
n_jobs (int) – The number of CPU cores to use. Defaults to -1.
random_seed (int) – The seed to use for random sampling. Defaults to 0.
Attributes
hyperparameter_ranges
{}
model_family
ModelFamily.NONE
modifies_features
True
modifies_target
True
name
SMOTE Oversampler
Methods
Constructs a new component with the same parameters and random state.
Returns the default parameters for this component.
Describe a component and its parameters
Fits the Oversampler to the data.
Fit and transform the data using the data sampler. Used during training of the pipeline
Loads component at file path
Returns boolean determining if component needs fitting before
Returns the parameters which were used to initialize the component
Saves component at file path
No transformation needs to be done here.
-
clone(self)¶ Constructs a new component with the same parameters and random state.
- Returns
A new instance of this component with identical parameters and random state.
-
default_parameters(cls)¶ Returns the default parameters for this component.
Our convention is that Component.default_parameters == Component().parameters.
- Returns
default parameters for this component.
- Return type
dict
-
describe(self, print_name=False, return_dict=False)¶ Describe a component and its parameters
- Parameters
print_name (bool, optional) – whether to print name of component
return_dict (bool, optional) – whether to return description as dictionary in the format {“name”: name, “parameters”: parameters}
- Returns
prints and returns dictionary
- Return type
None or dict
-
fit(self, X, y)¶ Fits the Oversampler to the data.
- Parameters
X (pd.DataFrame) – Training features
y (pd.Series) – Target features
- Returns
self
-
fit_transform(self, X, y)¶ Fit and transform the data using the data sampler. Used during training of the pipeline
- Parameters
X (pd.DataFrame) – Training features
y – Target features
-
static
load(file_path)¶ Loads component at file path
- Parameters
file_path (str) – Location to load file
- Returns
ComponentBase object
-
needs_fitting(self)¶ Returns boolean determining if component needs fitting before calling predict, predict_proba, transform, or feature_importances. This can be overridden to False for components that do not need to be fit or whose fit methods do nothing.
-
property
parameters(self)¶ Returns the parameters which were used to initialize the component
-
save(self, file_path, pickle_protocol=cloudpickle.DEFAULT_PROTOCOL)¶ Saves component at file path
- Parameters
file_path (str) – Location to save file
pickle_protocol (int) – The pickle data stream format.
- Returns
None
-
transform(self, X, y=None)¶ No transformation needs to be done here.
- Parameters
X (pd.DataFrame) – Training features. Ignored.
y (pd.Series) – Target features. Ignored.
- Returns
X and y data that was passed in.
- Return type
pd.DataFrame, pd.Series
-
class
evalml.pipelines.components.transformers.samplers.Undersampler(sampling_ratio=0.25, sampling_ratio_dict=None, min_samples=100, min_percentage=0.1, random_seed=0, **kwargs)[source]¶ Initializes an undersampling transformer to downsample the majority classes in the dataset.
This component is only run during training and not during predict.
- Parameters
sampling_ratio (float) – The smallest minority:majority ratio that is accepted as ‘balanced’. For instance, a 1:4 ratio would be represented as 0.25, while a 1:1 ratio is 1.0. Must be between 0 and 1, inclusive. Defaults to 0.25.
sampling_ratio_dict (dict) – A dictionary specifying the desired balanced ratio for each target value. For instance, in a binary case where class 1 is the minority, we could specify: sampling_ratio_dict={0: 0.5, 1: 1}, which means we would undersample class 0 to have twice the number of samples as class 1 (minority:majority ratio = 0.5), and don’t sample class 1. Overrides sampling_ratio if provided. Defaults to None.
min_samples (int) – The minimum number of samples that we must have for any class, pre or post sampling. If a class must be downsampled, it will not be downsampled past this value. To determine severe imbalance, the minority class must occur less often than this and must have a class ratio below min_percentage. Must be greater than 0. Defaults to 100.
min_percentage (float) – The minimum percentage of the minimum class to total dataset that we tolerate, as long as it is above min_samples. If min_percentage and min_samples are not met, treat this as severely imbalanced, and we will not resample the data. Must be between 0 and 0.5, inclusive. Defaults to 0.1.
random_seed (int) – The seed to use for random sampling. Defaults to 0.
Attributes
hyperparameter_ranges
{}
model_family
ModelFamily.NONE
modifies_features
True
modifies_target
True
name
Undersampler
Methods
Constructs a new component with the same parameters and random state.
Returns the default parameters for this component.
Describe a component and its parameters
Resample the data using the sampler. Since our sampler doesn’t need to be fit, we do nothing here.
Fit and transform the data using the undersampler. Used during training of the pipeline
Loads component at file path
Returns boolean determining if component needs fitting before
Returns the parameters which were used to initialize the component
Saves component at file path
No transformation needs to be done here.
-
clone(self)¶ Constructs a new component with the same parameters and random state.
- Returns
A new instance of this component with identical parameters and random state.
-
default_parameters(cls)¶ Returns the default parameters for this component.
Our convention is that Component.default_parameters == Component().parameters.
- Returns
default parameters for this component.
- Return type
dict
-
describe(self, print_name=False, return_dict=False)¶ Describe a component and its parameters
- Parameters
print_name (bool, optional) – whether to print name of component
return_dict (bool, optional) – whether to return description as dictionary in the format {“name”: name, “parameters”: parameters}
- Returns
prints and returns dictionary
- Return type
None or dict
-
fit(self, X, y)¶ Resample the data using the sampler. Since our sampler doesn’t need to be fit, we do nothing here.
- Parameters
X (pd.DataFrame) – Training features
y (pd.Series) – Target features
- Returns
self
-
fit_transform(self, X, y)[source]¶ Fit and transform the data using the undersampler. Used during training of the pipeline
- Parameters
X (pd.DataFrame) – Training features
y – Target features
-
static
load(file_path)¶ Loads component at file path
- Parameters
file_path (str) – Location to load file
- Returns
ComponentBase object
-
needs_fitting(self)¶ Returns boolean determining if component needs fitting before calling predict, predict_proba, transform, or feature_importances. This can be overridden to False for components that do not need to be fit or whose fit methods do nothing.
-
property
parameters(self)¶ Returns the parameters which were used to initialize the component
-
save(self, file_path, pickle_protocol=cloudpickle.DEFAULT_PROTOCOL)¶ Saves component at file path
- Parameters
file_path (str) – Location to save file
pickle_protocol (int) – The pickle data stream format.
- Returns
None
-
transform(self, X, y=None)¶ No transformation needs to be done here.
- Parameters
X (pd.DataFrame) – Training features. Ignored.
y (pd.Series) – Target features. Ignored.
- Returns
X and y data that was passed in.
- Return type
pd.DataFrame, pd.Series