TimeseriesAnomalyTrainingConfig

Training config for the TS_ANOMALY problem type

KEY TYPE Description
ANOMALY_TYPE TimeseriesAnomalyTypeOfAnomaly select what kind of peaks to detect as anomalies
HYPERPARAMETER_CALCULATION_WITH_HEURISTICS TimeseriesAnomalyUseHeuristic Enable heuristic calculation to get hyperparameters for the model
TIMESERIES_FREQUENCY str set this to control frequency of filling missing values
ADDITIONAL_ANOMALY_IDS List[str] List of categorical columns that can act as multi-identifier
TEST_SPLIT int Percent of dataset to use for test data. We support using a range between 5 ( i.e. 5% ) to 20 ( i.e. 20% ) of your dataset.
THRESHOLD_SCORE float Threshold score for anomaly detection
FILL_MISSING_VALUES List[List[dict]] strategies to fill missing values and missing timestamps
TYPE_OF_SPLIT TimeseriesAnomalyDataSplitType Type of data splitting into train/test.
MIN_SAMPLES_IN_NORMAL_REGION int Adjust this to fine-tune the number of anomalies to be identified.
TEST_START str Limit training data to dates before the given test start.
HANDLE_ZEROS_AS_MISSING_VALUES bool If True, handle zero values in numeric columns as missing data
© Copyright