SMOTEConverter

Converter

DashAI.back.converters.imbalanced_learn.SMOTEConverter

---

Balances class distribution by generating synthetic minority-class samples.

SMOTE (Synthetic Minority Over-sampling Technique) addresses class imbalance by creating new minority-class examples via linear interpolation between each minority sample and one of its k nearest minority-class neighbours. Unlike simple random over-sampling (which duplicates existing rows), SMOTE generates novel samples in the feature space, improving classifier generalisation.

The technique is applied only during training; the test split is never resampled. All schema parameters are forwarded to imbalanced-learn's SMOTE estimator.

References

• [1] Chawla, N.V. et al. (2002). "SMOTE: Synthetic Minority Over-sampling Technique." Journal of Artificial Intelligence Research, 16, 321-357. https://arxiv.org/abs/1106.1813
• [2] https://imbalanced-learn.org/stable/references/generated/imblearn.over_sampling.SMOTE.html

Parameters

sampling_strategy, default=auto

Sampling strategy (float or 'auto') to determine minority class size.

random_state, default=None

Seed for reproducibility.

k_neighbors : integer, default=5

Number of neighbors to use for generating synthetic samples.

Methods

get_output_type(self, column_name: str = None) -> DashAI.back.types.dashai_data_type.DashAIDataType

Defined on SMOTEConverter

Not implemented; type preservation is handled in transform.

Parameters

column_name : str or None, optional

Name of the column whose output type is queried. Ignored because this method always raises. Default None.

changes_row_count(self) -> bool

Defined on ImbalancedLearnWrapper

Return True because all samplers add or remove rows.

Returns

bool

Always True.

fit(self, x: 'DashAIDataset', y: 'DashAIDataset') -> Type[DashAI.back.converters.base_converter.BaseConverter]

Defined on ImbalancedLearnWrapper

Resample the dataset by calling fit_resample and store the result.

Parameters

x : DashAIDataset

The input feature dataset.

y : DashAIDataset

The target label dataset (required; must be non-empty).

Returns

Type[BaseConverter]

The fitted sampler instance (self).

get_metadata(cls) -> 'Dict[str, Any]'

Defined on BaseConverter

Get metadata for the converter, used by the DashAI frontend.

Parameters

cls : type

The converter class (injected automatically by Python for classmethods).

Returns

Dict[str, Any]

Dictionary containing display name, short description, image preview path, category, icon, color, and whether the converter is supervised.

get_schema(cls) -> dict

Defined on ConfigObject

Generates the component related Json Schema.

Returns

dict

Dictionary representing the Json Schema of the component.

transform(self, x: 'DashAIDataset', y: Optional[ForwardRef('DashAIDataset')] = None) -> 'DashAIDataset'

Defined on ImbalancedLearnWrapper

Return the resampled dataset stored during fit.

Parameters

x : DashAIDataset

The original feature dataset (used only for type information).

y : DashAIDataset, optional

The original target dataset (used only for type information). Defaults to None.

Returns

DashAIDataset

The combined resampled dataset (features + target) produced by fit.

validate_and_transform(self, raw_data: dict) -> dict

Defined on ConfigObject

It takes the data given by the user to initialize the model and returns it with all the objects that the model needs to work.

Parameters

raw_data : dict

A dictionary with the data provided by the user to initialize the model.

Returns

dict

A validated dictionary with the necessary objects.