pdmlabs.method.sklearn_regression_wraper

Classes

AdaBoostRUL(event_preferences[, save_model])
BaseSklearnRUL(event_preferences[, save_model])	Base class for sklearn regressors following the RUL interface.
BayesianRidgeRUL(event_preferences[, save_model])
DecisionTreeRUL(event_preferences[, save_model])
ElasticNetRUL(event_preferences[, save_model])
GPRRUL(event_preferences[, save_model])
GradientBoostingRUL(event_preferences[, ...])
KNNRUL(event_preferences[, save_model])
KernelRidgeRUL(event_preferences[, save_model])
LassoRUL(event_preferences[, save_model])
LinearRegressionRUL(event_preferences[, ...])
MLPRUL(event_preferences[, save_model])
RandomForestRUL(event_preferences[, save_model])
RidgeRUL(event_preferences[, save_model])
SVRRUL(event_preferences[, save_model])

class pdmlabs.method.sklearn_regression_wraper.AdaBoostRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of AdaBoostRegressor

class pdmlabs.method.sklearn_regression_wraper.BaseSklearnRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: SupervisedMethodInterface

Base class for sklearn regressors following the RUL interface.

MODEL_CLASS = 'sklearn'

fit(historic_data: list[DataFrame], historic_sources: list[str], event_data: DataFrame, anomaly_ranges: list[list]) → None

Train anomaly detection model on labeled data.

Fits the model to distinguish normal from anomalous behavior using provided training data and labels.

Parameters:

• historic_data (list[pd.DataFrame]) – List of training DataFrames, one per source. Each DataFrame has: - Columns: feature names (sensor readings, computed metrics) - Index: datetime (must be sorted) - Shape: (num_samples, num_features) All DataFrames should have same features.

• historic_sources (list[str]) – Source identifiers corresponding to historic_data. Example: [‘bearing_1’, ‘bearing_2’, ‘pump_1’]. Length must match len(historic_data).

• event_data (pd.DataFrame) – Event log for context. Columns should include ‘date’, ‘type’ (‘failure’, ‘reset’, etc.), ‘source’, and optional ‘description’. Can help training identify event patterns or validate training data selection.

• anomaly_ranges (list[list]) –

  Labels marking anomalous time periods. Structure: list of lists where element i corresponds to historic_sources[i]. Each element is list of time indices or ranges marking anomalies. Examples: - Index-based: [[10:50, 100:120], [5:30]] (indices start:end) - Boolean: [pd.Series([0,1,0,…]), pd.Series([0,0,1,…])]

  Use case: If bearing_1 has anomalies from idx 10-50, anomaly_ranges[0] includes that range.

Returns:

Modifies internal state to store fitted model.

Return type:

None

Raises:

• ValueError – If data shapes don’t match or labels invalid.

• NotImplementedError – Implementation not complete (abstract).

Notes

• Implementations may copy historic_data if they need to store it

• Training may take significant time for large datasets

• After fit(), predict() can be called

• Multiple fit() calls should retrain (not append)

Examples

>>> # Binary labels (0/1 for each sample)
>>> labels = [np.array([0, 0, 1, 1, 0, ...]), np.array([0, 0, 0, ...])]
>>> method.fit(data, sources, events, labels)
>>>
>>> # Time ranges (list of anomalous periods)
>>> anomaly_periods = [
...     [(start_idx1, end_idx1), (start_idx2, end_idx2)],
...     [(start_idx3, end_idx3)],
... ]
>>> method.fit(data, sources, events, anomaly_periods)

get_all_models()

Return reference to internal model(s).

Returns model instances for inspection, export, or further processing. Structure depends on method - may return single model or dict of models.

Returns:

Underlying model object(s). Examples:

• Single model: sklLearn model instance

• Multiple models: {‘bearing_1’: model1, ‘bearing_2’: model2}

• None: If model is not accessible/applicable

Return type:

model or dict

get_library() → str

Return the underlying library/framework name.

Returns:

Name of library used (e.g., ‘sklearn’, ‘torch’, ‘custom’).

Used for dependency tracking and method categorization.

Return type:

str

get_params() → dict

Return hyperparameters and configuration.

Returns:

Dictionary of hyperparameters (e.g.,

{‘n_neighbors’: 5, ‘contamination’: 0.1}). Useful for logging, model comparison, and reproducibility.

Return type:

dict

predict(target_data: DataFrame, source: str, event_data: DataFrame) → list[float]

Predict anomaly scores for batch of samples (offline mode).

Computes anomaly score for each row in target_data independently. Higher scores indicate more anomalous behavior.

Parameters:

• target_data (pd.DataFrame) – Feature matrix with dates in index, features in columns. Must have same features as training data.

• source (str) – Source identifier (e.g., ‘bearing_1’). Used to select source-specific model if method maintains multiple models.

• event_data (pd.DataFrame) – Event log with columns ‘date’, ‘type’, ‘source’, ‘description’. Can be used for context-aware scoring.

Returns:

Anomaly scores (float) with length = target_data.shape[0].

Score range and semantics depend on method: - Distance-based: typically [0, ∞) where higher = more anomalous - Probability-based: typically [0, 1] or (-∞, 0] log-likelihood - Reconstruction-based: typically [0, ∞) reconstruction error

Return type:

list

Examples

>>> method = SomeAnomalyDetector(event_preferences={...})
>>> method.fit([df_train], ['bearing_1'], events_df, labels)
>>> df_test = pd.DataFrame([feature values], index=[dates])
>>> scores = method.predict(df_test, 'bearing_1', events_df)
>>> print(len(scores), scores[0])  # (100, 0.45)

Raises:

NotImplementedError – If method hasn’t been fit (for supervised methods).

predict_one(new_sample: Series, source: str, is_event: bool) → float

Predict anomaly score for single sample (online/streaming mode).

Computes anomaly score for one observation at a time. Useful for: - Real-time anomaly detection - Online learning scenarios - Memory-efficient processing

May maintain internal state (buffers, windows) for context-aware scoring.

Parameters:

• new_sample (pd.Series) – Single observation with feature values. Index should contain feature names matching training data.

• source (str) – Source identifier for source-specific models.

• is_event (bool) – Whether this sample is from an event timestamp. Can affect how method processes the sample (e.g., special handling for known events vs normal operation).

Returns:

Anomaly score for this single sample (same scale as predict()).

Return type:

float

Examples

>>> method = SomeAnomalyDetector(event_preferences={...})
>>> method.fit([df_train], ['bearing_1'], events_df, labels)
>>>
>>> # Online scoring
>>> for idx, row in df_test.iterrows():
...     is_event = idx in event_timestamps
...     score = method.predict_one(row, 'bearing_1', is_event)
...     print(f'{idx}: {score}')

class pdmlabs.method.sklearn_regression_wraper.BayesianRidgeRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of BayesianRidge

class pdmlabs.method.sklearn_regression_wraper.DecisionTreeRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of DecisionTreeRegressor

class pdmlabs.method.sklearn_regression_wraper.ElasticNetRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of ElasticNet

class pdmlabs.method.sklearn_regression_wraper.GPRRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of GaussianProcessRegressor

class pdmlabs.method.sklearn_regression_wraper.GradientBoostingRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of GradientBoostingRegressor

class pdmlabs.method.sklearn_regression_wraper.KNNRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of KNeighborsRegressor

class pdmlabs.method.sklearn_regression_wraper.KernelRidgeRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of KernelRidge

class pdmlabs.method.sklearn_regression_wraper.LassoRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of Lasso

class pdmlabs.method.sklearn_regression_wraper.LinearRegressionRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of LinearRegression

class pdmlabs.method.sklearn_regression_wraper.MLPRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of MLPRegressor

class pdmlabs.method.sklearn_regression_wraper.RandomForestRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of RandomForestRegressor

class pdmlabs.method.sklearn_regression_wraper.RidgeRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of Ridge

class pdmlabs.method.sklearn_regression_wraper.SVRRUL(event_preferences: EventPreferences, save_model=False, *args, **kwargs)

Bases: BaseSklearnRUL

MODEL_CLASS

alias of SVR