From a7644438c2ee90dc7b8bc77fe831941a0438e2c9 Mon Sep 17 00:00:00 2001
From: Carlosbogo <84228424+Carlosbogo@users.noreply.github.com>
Date: Sat, 27 Nov 2021 21:03:07 +0100
Subject: [PATCH] Closes #290 (#313)

* Closes #290
Renames parameters for number of folds
in different functions to keep consistency
---
 etna/ensembles/stacking_ensemble.py            |  8 ++++----
 etna/pipeline/pipeline.py                      | 10 +++++-----
 tests/test_ensembles/test_stacking_ensemble.py |  4 ++--
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/etna/ensembles/stacking_ensemble.py b/etna/ensembles/stacking_ensemble.py
index fa82286dd..e639e666c 100644
--- a/etna/ensembles/stacking_ensemble.py
+++ b/etna/ensembles/stacking_ensemble.py
@@ -62,7 +62,7 @@ def __init__(
         self,
         pipelines: List[Pipeline],
         final_model: RegressorMixin = LinearRegression(),
-        cv: int = 3,
+        n_folds: int = 3,
         features_to_use: Union[None, Literal["all"], List[str]] = None,
         n_jobs: int = 1,
         joblib_params: Dict[str, Any] = dict(verbose=11, backend="multiprocessing", mmap_mode="c"),
@@ -75,7 +75,7 @@ def __init__(
             List of pipelines that should be used in ensemble.
         final_model:
             Regression model with fit/predict interface which will be used to combine the base estimators.
-        cv:
+        n_folds:
             Number of folds to use in the backtest. Backtest is not used for model evaluation but for prediction.
         features_to_use:
             Features except the forecasts of the base models to use in the `final_model`.
@@ -93,7 +93,7 @@ def __init__(
         self.pipelines = pipelines
         self.horizon = self._get_horizon(pipelines=pipelines)
         self.final_model = final_model
-        self.cv = self._validate_cv(cv)
+        self.n_folds = self._validate_cv(n_folds)
         self.features_to_use = features_to_use
         self.filtered_features_for_final_model: Union[None, Set[str]] = None
         self.n_jobs = n_jobs
@@ -149,7 +149,7 @@ def _fit_pipeline(pipeline: Pipeline, ts: TSDataset) -> Pipeline:
 
     def _backtest_pipeline(self, pipeline: Pipeline, ts: TSDataset) -> TSDataset:
         """Get forecasts from backtest for given pipeline."""
-        _, forecasts, _ = pipeline.backtest(ts, metrics=[MAE()], n_folds=self.cv)
+        _, forecasts, _ = pipeline.backtest(ts, metrics=[MAE()], n_folds=self.n_folds)
         forecasts = TSDataset(df=forecasts, freq=ts.freq)
         return forecasts
 
diff --git a/etna/pipeline/pipeline.py b/etna/pipeline/pipeline.py
index f5634bd26..689d7526f 100644
--- a/etna/pipeline/pipeline.py
+++ b/etna/pipeline/pipeline.py
@@ -43,7 +43,7 @@ def __init__(
         transforms: Sequence[Transform] = (),
         horizon: int = 1,
         quantiles: Sequence[float] = (0.025, 0.975),
-        prediction_interval_cv: int = 3,
+        n_folds: int = 3,
     ):
         """
         Create instance of Pipeline with given parameters.
@@ -58,19 +58,19 @@ def __init__(
             Number of timestamps in the future for forecasting
         quantiles:
             Levels of prediction distribution. By default 2.5% and 97.5% taken to form a 95% prediction interval
-        prediction_interval_cv:
+        n_folds:
             Number of folds to use in the backtest for prediction interval estimation
 
         Raises
         ------
         ValueError:
-            If the horizon is less than 1, quantile is out of (0,1) or prediction_interval_cv is less than 2.
+            If the horizon is less than 1, quantile is out of (0,1) or n_folds is less than 2.
         """
         super().__init__(quantiles=quantiles)
         self.model = model
         self.transforms = transforms
         self.horizon = self._validate_horizon(horizon)
-        self.prediction_interval_cv = self._validate_cv(prediction_interval_cv)
+        self.n_folds = self._validate_cv(n_folds)
         self.ts: Optional[TSDataset] = None
 
     @staticmethod
@@ -111,7 +111,7 @@ def fit(self, ts: TSDataset) -> "Pipeline":
 
     def _forecast_prediction_interval(self, future: TSDataset) -> TSDataset:
         """Forecast prediction interval for the future."""
-        _, forecasts, _ = self.backtest(self.ts, metrics=[MAE()], n_folds=self.prediction_interval_cv)
+        _, forecasts, _ = self.backtest(self.ts, metrics=[MAE()], n_folds=self.n_folds)
         forecasts = TSDataset(df=forecasts, freq=self.ts.freq)
         residuals = (
             forecasts.loc[:, pd.IndexSlice[:, "target"]]
diff --git a/tests/test_ensembles/test_stacking_ensemble.py b/tests/test_ensembles/test_stacking_ensemble.py
index e0e5ee01f..2c37d7ef1 100644
--- a/tests/test_ensembles/test_stacking_ensemble.py
+++ b/tests/test_ensembles/test_stacking_ensemble.py
@@ -38,7 +38,7 @@ def test_get_horizon_fail(catboost_pipeline: Pipeline, naive_pipeline: Pipeline)
 @pytest.mark.parametrize("input_cv,true_cv", ([(2, 2)]))
 def test_cv_pass(naive_pipeline_1: Pipeline, naive_pipeline_2: Pipeline, input_cv, true_cv):
     """Check that StackingEnsemble._validate_cv works correctly in case of valid cv parameter."""
-    ensemble = StackingEnsemble(pipelines=[naive_pipeline_1, naive_pipeline_2], cv=input_cv)
+    ensemble = StackingEnsemble(pipelines=[naive_pipeline_1, naive_pipeline_2], n_folds=input_cv)
     assert ensemble.cv == true_cv
 
 
@@ -46,7 +46,7 @@ def test_cv_pass(naive_pipeline_1: Pipeline, naive_pipeline_2: Pipeline, input_c
 def test_cv_fail_wrong_number(naive_pipeline_1: Pipeline, naive_pipeline_2: Pipeline, input_cv):
     """Check that StackingEnsemble._validate_cv works correctly in case of wrong number for cv parameter."""
     with pytest.raises(ValueError, match="At least two folds for backtest are expected."):
-        _ = StackingEnsemble(pipelines=[naive_pipeline_1, naive_pipeline_2], cv=input_cv)
+        _ = StackingEnsemble(pipelines=[naive_pipeline_1, naive_pipeline_2], n_folds=input_cv)
 
 
 @pytest.mark.parametrize(