Verifying the performance of the bounded detectors on an univariate example

TODO: improve these tests to introduce some correlation between actual and detected anomalies

If two values are given for low and high, and a series of width 2 is given, then the results must be the same as a detector that was given onl

This test checks whether calling the scaler twice is calculating new lambdas instead of keeping the old ones

Test calibration historical forecast start point when calling `historical_forecasts()` with start format "position".

Test calibration historical forecast start point when calling `predict()` ("end" position).

Test calibration historical forecast start point when calling `historical_forecasts()` with start format "value".

Test `CallableIndexEncoder`

Test `CallableIndexEncoder` with a callable returning multiple components.

Test whether the categorical features are passed to fit correctly

Check class probabilities have correct shape and meaning in case of: - single series, univariate - single series, multivariat

Tests if the `columns` argument correctly applies the transform only to the specified columns.

Tests if the `columns` argument correctly applies the transform and it's inverse only to the specified columns.

Tests if the `columns` argument correctly applies the transform and it's inverse only to the specified columns.

Tests if the `columns` argument correctly applies the transform only to the specified columns.

Tests if monthly series aligned with quarters is transformed into a quarterly series in the expected way.

Checking various combination of component-specific lags

Verify that the same lags, defined using int/list or dictionaries yield the same results, including output_chunk_shift.

test concatenation with static covariates along component dimension (axis=1) Along component dimension, we concatenate/transfer the s

Test concatenation with static covariates along sample dimension (axis=2) Along sample dimension, we only take the static covariates

Test concatenation with static covariates along time dimension (axis=0) Along time dimension, we only take the static covariates of t

Test that conditional seasonality is correctly incorporated by the model

Tests naive conformal model with `start` being the first forecastable index is identical to a start before forecastable index (including strid

Tests historical forecasts output naive conformal model with last points only, covariates, stride, different horizons and overlap end.

Tests naive conformal model historical forecasts without `cal_stride`.

Verifies that naive conformal model computes the correct intervals for: - different horizons (smaller, equal, larger than ocl) - uni/m

Using kwargs to pass additional parameters to the constructor

Tests built in weights generation for: - varying target series sizes - with and without moving window tabularization - differe

Checks correct weights extraction for: - varying target series sizes - with and without moving window tabularization - weights

Tests that `create_lagged_component_names` produces the expected features name depending on the lags, output_chunk_length and covaria

Tests that `create_lagged_component_names` when lags are different across components. The lagged features should be sorted by lags,