nannyml.drift.multivariate.data_reconstruction.result module¶
Contains the results of the data reconstruction drift calculation and provides plotting functionality.
- class nannyml.drift.multivariate.data_reconstruction.result.Metric(display_name, column_name)¶
Bases:
tuple
Create new instance of Metric(display_name, column_name)
- column_name¶
Alias for field number 1
- display_name¶
Alias for field number 0
- class nannyml.drift.multivariate.data_reconstruction.result.Result(results_data: pandas.core.frame.DataFrame, column_names: List[str], categorical_column_names: List[str], continuous_column_names: List[str], timestamp_column_name: Optional[str] = None)[source]¶
Bases:
nannyml.base.Abstract1DResult
,nannyml.plots.blueprints.comparisons.ResultCompareMixin
Contains the results of the data reconstruction drift calculation and provides plotting functionality.
Creates a new
AbstractCalculatorResult
instance.- Parameters
results_data (pd.DataFrame) – The data returned by the Calculator.
- plot(kind: str = 'drift', *args, **kwargs) Optional[plotly.graph_objs._figure.Figure] [source]¶
Renders plots for metrics returned by the multivariate data reconstruction calculator.
The different plot kinds that are available:
drift
plots the multivariate reconstruction error over the provided features per
Chunk
.
- Parameters
kind (str, default=`drift`) – The kind of plot you want to have. Value can currently only be
drift
.plot_reference (bool, default=False) – Indicates whether to include the reference period in the plot or not. Defaults to
False
.
- Returns
fig – A
Figure
object containing the requested drift plot.Can be saved to disk using the
write_image()
method or shown rendered on screen using theshow()
method.- Return type
plotly.graph_objs._figure.Figure
Examples
>>> import nannyml as nml >>> reference_df, analysis_df, _ = nml.load_synthetic_binary_classification_dataset() >>> >>> column_names = [col for col in reference_df.columns >>> if col not in ['y_pred', 'y_pred_proba', 'work_home_actual', 'timestamp']] >>> calc = nml.DataReconstructionDriftCalculator( >>> column_names=column_names, >>> timestamp_column_name='timestamp' >>> ) >>> calc.fit(reference_df) >>> results = calc.calculate(analysis_df) >>> print(results.data) # access the numbers key start_index ... upper_threshold alert 0 [0:4999] 0 ... 1.511762 True 1 [5000:9999] 5000 ... 1.511762 True 2 [10000:14999] 10000 ... 1.511762 True 3 [15000:19999] 15000 ... 1.511762 True 4 [20000:24999] 20000 ... 1.511762 True 5 [25000:29999] 25000 ... 1.511762 True 6 [30000:34999] 30000 ... 1.511762 True 7 [35000:39999] 35000 ... 1.511762 True 8 [40000:44999] 40000 ... 1.511762 True 9 [45000:49999] 45000 ... 1.511762 True >>> fig = results.plot(plot_reference=True) >>> fig.show()