Fix WindowStatisticsTransform (#1128)

etna/transforms/encoders/categorical.py

@@ -1,4 +1,3 @@
-import warnings
 from enum import Enum
 from typing import List
 from typing import Optional
@@ -78,7 +77,7 @@ def __init__(self, in_column: str, out_column: Optional[str] = None, strategy: s
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
         if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
         return [self._get_column_name()] if self.in_column_regressor else []
 
     def _fit(self, df: pd.DataFrame) -> "LabelEncoderTransform":
@@ -159,7 +158,7 @@ def __init__(self, in_column: str, out_column: Optional[str] = None):
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
         if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
         return self._get_out_column_names() if self.in_column_regressor else []
 
     def _fit(self, df: pd.DataFrame) -> "OneHotEncoderTransform":

etna/transforms/math/add_constant.py

@@ -136,7 +136,7 @@ def _inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
         if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
         return [self._get_column_name()] if self.in_column_regressor and not self.inplace else []
 
 

etna/transforms/math/apply_lambda.py

@@ -154,5 +154,5 @@ def _inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
         if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
         return [self.change_column] if self.in_column_regressor and not self.inplace else []

etna/transforms/math/differencing.py

@@ -1,4 +1,3 @@
-import warnings
 from typing import Dict
 from typing import List
 from typing import Optional
@@ -81,10 +80,10 @@ def _get_column_name(self) -> str:
 
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
+        if self.in_column_regressor is None:
+            raise ValueError("Fit the transform to get the correct regressors info!")
         if self.inplace:
             return []
-        if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
         return [self._get_column_name()] if self.in_column_regressor else []
 
     def fit(self, ts: TSDataset) -> "_SingleDifferencingTransform":
@@ -359,10 +358,10 @@ def _get_column_name(self) -> str:
 
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
+        if self.in_column_regressor is None:
+            raise ValueError("Fit the transform to get the correct regressors info!")
         if self.inplace:
             return []
-        if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
         return [self._get_column_name()] if self.in_column_regressor else []
 
     def fit(self, ts: TSDataset) -> "DifferencingTransform":

etna/transforms/math/log.py

@@ -139,7 +139,7 @@ def _inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
         if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
         return [self._get_column_name()] if self.in_column_regressor and not self.inplace else []
 
 

etna/transforms/math/sklearn.py

@@ -251,9 +251,8 @@ def _inverse_reshape(self, df: pd.DataFrame, transformed: np.ndarray) -> np.ndar
 
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
-        if self.inplace:
-            return []
         if self.out_column_regressors is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
+            raise ValueError("Fit the transform to get the correct regressors info!")
+        if self.inplace:
             return []
         return self.out_column_regressors

etna/transforms/math/statistics.py

@@ -7,6 +7,7 @@
 import numpy as np
 import pandas as pd
 
+from etna.datasets import TSDataset
 from etna.transforms.base import IrreversibleTransform
 
 
@@ -49,6 +50,13 @@ def __init__(
         self.min_periods = min_periods
         self.fillna = fillna
         self.kwargs = kwargs
+        self.in_column_regressor: Optional[bool] = None
+
+    def fit(self, ts: TSDataset) -> "WindowStatisticsTransform":
+        """Fit the transform."""
+        self.in_column_regressor = self.in_column in ts.regressors
+        super().fit(ts)
+        return self
 
     def _fit(self, df: pd.DataFrame) -> "WindowStatisticsTransform":
         """Fits transform."""
@@ -100,7 +108,9 @@ def _transform(self, df: pd.DataFrame) -> pd.DataFrame:
 
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
-        return [self.out_column_name]
+        if self.in_column_regressor is None:
+            raise ValueError("Fit the transform to get the correct regressors info!")
+        return [self.out_column_name] if self.in_column_regressor else []
 
 
 class MeanTransform(WindowStatisticsTransform):

etna/transforms/missing_values/resample.py

@@ -164,10 +164,10 @@ def _get_out_column(self, out_column: Optional[str]) -> str:
 
     def get_regressors_info(self) -> List[str]:
         """Return the list with regressors created by the transform."""
+        if self.in_column_regressor is None:
+            raise ValueError("Fit the transform to get the correct regressors info!")
         if self.inplace:
             return []
-        if self.in_column_regressor is None:
-            warnings.warn("Regressors info might be incorrect. Fit the transform to get the correct regressors info.")
         return [self.out_column] if self.in_column_regressor else []
 
     def fit(self, ts: TSDataset) -> "ResampleWithDistributionTransform":

tests/test_transforms/test_encoders/test_categorical_transform.py

@@ -342,3 +342,11 @@ def test_save_load_ohe(dtype):
     for i in range(3):
         transform = OneHotEncoderTransform(in_column=f"regressor_{i}", out_column="test")
         assert_transformation_equals_loaded_original(transform=transform, ts=ts)
+
+
+@pytest.mark.parametrize(
+    "transform", (OneHotEncoderTransform(in_column="regressor_0"), LabelEncoderTransform(in_column="regressor_0"))
+)
+def test_get_regressors_info_not_fitted(transform):
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_feature_selection/test_gale_shapley_transform.py

@@ -654,8 +654,8 @@ def test_right_number_features_with_integer_division(ts_with_exog_galeshapley):
     top_k = len(ts_with_exog_galeshapley.segments)
     transform = GaleShapleyFeatureSelectionTransform(relevance_table=StatisticsRelevanceTable(), top_k=top_k)
 
-    transform.fit(ts_with_exog_galeshapley.to_pandas())
-    df = transform.transform(ts_with_exog_galeshapley.to_pandas())
+    transform.fit(ts_with_exog_galeshapley)
+    ts = transform.transform(ts_with_exog_galeshapley)
 
-    remaining_columns = df.columns.get_level_values("feature").unique().tolist()
+    remaining_columns = ts.columns.get_level_values("feature").unique().tolist()
     assert len(remaining_columns) == top_k + 1

tests/test_transforms/test_math/test_add_constant_transform.py

@@ -80,3 +80,9 @@ def test_fit_transform_with_nans(ts_diff_endings):
 def test_save_load(inplace, example_tsds):
     transform = AddConstTransform(in_column="target", value=10, inplace=inplace)
     assert_transformation_equals_loaded_original(transform=transform, ts=example_tsds)
+
+
+def test_get_regressors_info_not_fitted():
+    transform = AddConstTransform(in_column="target", value=1, out_column="out_column")
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_math/test_differencing_transform.py

@@ -474,3 +474,9 @@ def test_save_load(inplace, ts_nans):
     ts = ts_nans
     transform = DifferencingTransform(in_column="target", inplace=inplace)
     assert_transformation_equals_loaded_original(transform=transform, ts=ts)
+
+
+def test_get_regressors_info_not_fitted():
+    transform = DifferencingTransform(in_column="target")
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_math/test_lambda_transform.py

@@ -158,3 +158,9 @@ def test_save_load(inplace, ts_range_const):
         inverse_transform_func=example_inverse_transform_func,
     )
     assert_transformation_equals_loaded_original(transform=transform, ts=ts_range_const)
+
+
+def test_get_regressors_info_not_fitted():
+    transform = LambdaTransform(in_column="target", inplace=False, transform_func=lambda x: x)
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_math/test_log_transform.py

@@ -118,3 +118,9 @@ def test_save_load(inplace, positive_ts_):
     ts = positive_ts_
     transform = LogTransform(in_column="target", inplace=inplace)
     assert_transformation_equals_loaded_original(transform=transform, ts=ts)
+
+
+def test_get_regressors_info_not_fitted():
+    transform = LogTransform(in_column="target")
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_math/test_sklearn_transform_interface.py

@@ -277,3 +277,23 @@ def test_ordering(transform_constructor, in_column, mode, multicolumn_ts):
         df_multi = transformed_df.loc[:, pd.IndexSlice[segments, column_multi]]
         df_single = transformed_dfs_one_column[i].loc[:, pd.IndexSlice[segments, column_single]]
         assert np.all(df_multi.values == df_single.values)
+
+
+@pytest.mark.parametrize(
+    "transform_constructor",
+    [
+        BoxCoxTransform,
+        YeoJohnsonTransform,
+        StandardScalerTransform,
+        RobustScalerTransform,
+        MinMaxScalerTransform,
+        MaxAbsScalerTransform,
+        StandardScalerTransform,
+        RobustScalerTransform,
+        MinMaxScalerTransform,
+    ],
+)
+def test_get_regressors_info_not_fitted(transform_constructor):
+    transform = transform_constructor(in_column="target")
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()

tests/test_transforms/test_math/test_statistics_transform.py

@@ -14,6 +14,7 @@
 from etna.transforms.math import QuantileTransform
 from etna.transforms.math import StdTransform
 from etna.transforms.math import SumTransform
+from etna.transforms.math import WindowStatisticsTransform
 from tests.test_transforms.utils import assert_transformation_equals_loaded_original
 
 
@@ -50,6 +51,25 @@ def ts_for_agg_with_nan() -> TSDataset:
     return ts
 
 
+class DummyWindowStatisticsTransform(WindowStatisticsTransform):
+    def _aggregate(self, series: np.ndarray):
+        return None
+
+
+@pytest.mark.parametrize("in_column,expected_regressors", (("target", []), ("regressor_exog_weekend", ["out_column"])))
+def test_get_regressors_info(example_reg_tsds, in_column, expected_regressors):
+    transform = DummyWindowStatisticsTransform(in_column=in_column, out_column="out_column", window=1)
+    transform.fit(ts=example_reg_tsds)
+    out_regressors = transform.get_regressors_info()
+    assert out_regressors == expected_regressors
+
+
+def test_get_regressors_info_not_fitted():
+    transform = DummyWindowStatisticsTransform(in_column="target", out_column="out_column", window=1)
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()
+
+
 @pytest.mark.parametrize(
     "class_name,out_column",
     (

tests/test_transforms/test_missing_values/test_resample_transform.py

@@ -121,3 +121,9 @@ def test_save_load(inplace, out_column, daily_exog_ts):
         in_column="regressor_exog", inplace=inplace, distribution_column="target", out_column=out_column
     )
     assert_transformation_equals_loaded_original(transform=transform, ts=daily_exog_ts)
+
+
+def test_get_regressors_info_not_fitted():
+    transform = ResampleWithDistributionTransform(in_column="regressor_exog", distribution_column="target")
+    with pytest.raises(ValueError, match="Fit the transform to get the correct regressors info!"):
+        _ = transform.get_regressors_info()