Add OneSegmentTransform (#894)

etna/transforms/__init__.py

@@ -2,6 +2,7 @@
 from etna.transforms.base import IrreversibleTransform
 from etna.transforms.base import NewPerSegmentWrapper
 from etna.transforms.base import NewTransform
+from etna.transforms.base import OneSegmentTransform
 from etna.transforms.base import PerSegmentWrapper
 from etna.transforms.base import ReversiblePerSegmentWrapper
 from etna.transforms.base import ReversibleTransform

etna/transforms/base.py

@@ -342,33 +342,103 @@ def inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
         return df
 
 
+class OneSegmentTransform(ABC, BaseMixin):
+    """Base class to create one segment transforms to apply to data."""
+
+    @abstractmethod
+    def fit(self, df: pd.DataFrame):
+        """Fit the transform.
+
+        Should be implemented by user.
+
+        Parameters
+        ----------
+        df:
+            Dataframe in etna long format.
+        """
+        pass
+
+    @abstractmethod
+    def transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """Transform dataframe.
+
+        Should be implemented by user
+
+        Parameters
+        ----------
+        df:
+            Dataframe in etna long format.
+
+        Returns
+        -------
+        :
+            Transformed Dataframe in etna long format.
+        """
+        pass
+
+    def fit_transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """Fit and transform Dataframe.
+
+        May be reimplemented. But it is not recommended.
+
+        Parameters
+        ----------
+        df:
+            Dataframe in etna long format to transform.
+
+        Returns
+        -------
+        :
+            Transformed Dataframe.
+        """
+        return self.fit(df=df).transform(df=df)
+
+    @abstractmethod
+    def inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """Inverse transform Dataframe.
+
+        Should be reimplemented in the subclasses where necessary.
+
+        Parameters
+        ----------
+        df:
+            Dataframe in etna long format to be inverse transformed.
+
+        Returns
+        -------
+        :
+            Dataframe after applying inverse transformation.
+        """
+        pass
+
+
 class NewPerSegmentWrapper(NewTransform):
     """Class to apply transform in per segment manner."""
 
-    def __init__(self, transform: NewTransform):
+    def __init__(self, transform: OneSegmentTransform, required_features: Union[Literal["all"], List[str]]):
         self._base_transform = transform
-        self.segment_transforms: Optional[Dict[str, NewTransform]] = None
-        super().__init__(required_features=transform.required_features)
+        self.segment_transforms: Optional[Dict[str, OneSegmentTransform]] = None
+        super().__init__(required_features=required_features)
 
     def _fit(self, df: pd.DataFrame):
         """Fit transform on each segment."""
         self.segment_transforms = {}
         segments = df.columns.get_level_values("segment").unique()
         for segment in segments:
             self.segment_transforms[segment] = deepcopy(self._base_transform)
-            self.segment_transforms[segment]._fit(df[segment])
+            self.segment_transforms[segment].fit(df[segment])
 
     def _transform(self, df: pd.DataFrame) -> pd.DataFrame:
         """Apply transform to each segment separately."""
         if self.segment_transforms is None:
             raise ValueError("Transform is not fitted!")
 
         results = []
-        for key, value in self.segment_transforms.items():
-            seg_df = value._transform(df[key])
+        for segment, transform in self.segment_transforms.items():
+            seg_df = transform.transform(df[segment])
 
             _idx = seg_df.columns.to_frame()
-            _idx.insert(0, "segment", key)
+            _idx.insert(0, "segment", segment)
             seg_df.columns = pd.MultiIndex.from_frame(_idx)
 
             results.append(seg_df)
@@ -382,27 +452,27 @@ def _transform(self, df: pd.DataFrame) -> pd.DataFrame:
 class IrreversiblePerSegmentWrapper(NewPerSegmentWrapper, IrreversibleTransform):
     """Class to apply irreversible transform in per segment manner."""
 
-    def __init__(self, transform: IrreversibleTransform):
-        super().__init__(transform=transform)
+    def __init__(self, transform: OneSegmentTransform, required_features: Union[Literal["all"], List[str]]):
+        super().__init__(transform=transform, required_features=required_features)
 
 
 class ReversiblePerSegmentWrapper(NewPerSegmentWrapper, ReversibleTransform):
     """Class to apply reversible transform in per segment manner."""
 
-    def __init__(self, transform: ReversibleTransform):
-        super().__init__(transform=transform)
+    def __init__(self, transform: OneSegmentTransform, required_features: Union[Literal["all"], List[str]]):
+        super().__init__(transform=transform, required_features=required_features)
 
     def _inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
         """Apply inverse_transform to each segment."""
         if self.segment_transforms is None:
             raise ValueError("Transform is not fitted!")
 
         results = []
-        for key, value in self.segment_transforms.items():
-            seg_df = value._inverse_transform(df[key])  # type: ignore
+        for segment, transform in self.segment_transforms.items():
+            seg_df = transform.inverse_transform(df[segment])
 
             _idx = seg_df.columns.to_frame()
-            _idx.insert(0, "segment", key)
+            _idx.insert(0, "segment", segment)
             seg_df.columns = pd.MultiIndex.from_frame(_idx)
 
             results.append(seg_df)

etna/transforms/missing_values/resample.py

@@ -4,14 +4,15 @@
 
 import pandas as pd
 
-from etna.transforms.base import PerSegmentWrapper
-from etna.transforms.base import Transform
+from etna.datasets import TSDataset
+from etna.transforms.base import IrreversiblePerSegmentWrapper
+from etna.transforms.base import OneSegmentTransform
 
 
-class _OneSegmentResampleWithDistributionTransform(Transform):
+class _OneSegmentResampleWithDistributionTransform(OneSegmentTransform):
     """_OneSegmentResampleWithDistributionTransform resamples the given column using the distribution of the other column."""
 
-    def __init__(self, in_column: str, distribution_column: str, inplace: bool, out_column: Optional[str]):
+    def __init__(self, in_column: str, distribution_column: str, inplace: bool, out_column: str):
         """
         Init _OneSegmentResampleWithDistributionTransform.
 
@@ -34,7 +35,7 @@ def __init__(self, in_column: str, distribution_column: str, inplace: bool, out_
         self.distribution_column = distribution_column
         self.inplace = inplace
         self.out_column = out_column
-        self.distribution: pd.DataFrame = None
+        self.distribution: Optional[pd.DataFrame] = None
 
     def _get_folds(self, df: pd.DataFrame) -> List[int]:
         """
@@ -101,8 +102,12 @@ def transform(self, df: pd.DataFrame) -> pd.DataFrame:
         df = df.drop(["fold", "distribution"], axis=1)
         return df
 
+    def inverse_transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """Inverse transform Dataframe."""
+        return df
+
 
-class ResampleWithDistributionTransform(PerSegmentWrapper):
+class ResampleWithDistributionTransform(IrreversiblePerSegmentWrapper):
     """ResampleWithDistributionTransform resamples the given column using the distribution of the other column.
 
     Warning
@@ -136,13 +141,15 @@ def __init__(
         self.distribution_column = distribution_column
         self.inplace = inplace
         self.out_column = self._get_out_column(out_column)
+        self.in_column_regressor: Optional[bool] = None
         super().__init__(
             transform=_OneSegmentResampleWithDistributionTransform(
                 in_column=in_column,
                 distribution_column=distribution_column,
                 inplace=inplace,
                 out_column=self.out_column,
-            )
+            ),
+            required_features=[in_column, distribution_column],
         )
 
     def _get_out_column(self, out_column: Optional[str]) -> str:
@@ -154,3 +161,17 @@ def _get_out_column(self, out_column: Optional[str]) -> str:
         if out_column:
             return out_column
         return self.__repr__()
+
+    def get_regressors_info(self) -> List[str]:
+        """Return the list with regressors created by the transform."""
+        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":
+        """Fit the transform."""
+        self.in_column_regressor = self.in_column in ts.regressors
+        super().fit(ts)
+        return self

tests/test_transforms/test_missing_values/test_resample_transform.py

@@ -8,7 +8,7 @@ def test_fail_on_incompatible_freq(incompatible_freq_ts):
         in_column="exog", inplace=True, distribution_column="target", out_column=None
     )
     with pytest.raises(ValueError, match="Can not infer in_column frequency!"):
-        _ = resampler.fit(incompatible_freq_ts.df)
+        _ = resampler.fit(incompatible_freq_ts)
 
 
 @pytest.mark.parametrize(
@@ -27,7 +27,7 @@ def test_fit(ts, request):
     resampler = ResampleWithDistributionTransform(
         in_column="regressor_exog", inplace=True, distribution_column="target", out_column=None
     )
-    resampler.fit(ts.df)
+    resampler.fit(ts)
     segments = ts.df.columns.get_level_values("segment").unique()
     for segment in segments:
         assert (resampler.segment_transforms[segment].distribution == expected_distribution[segment]).all().all()
@@ -48,10 +48,11 @@ def test_transform(daily_exog_ts, inplace, out_column, expected_resampled_ts, re
     resampler = ResampleWithDistributionTransform(
         in_column="regressor_exog", inplace=inplace, distribution_column="target", out_column=out_column
     )
-    resampled_df = resampler.fit_transform(daily_exog_ts.df)
+    resampled_df = resampler.fit_transform(daily_exog_ts).to_pandas()
     assert resampled_df.equals(expected_resampled_df)
 
 
+@pytest.mark.xfail(reason="TSDataset 2.0")
 @pytest.mark.parametrize(
     "inplace,out_column,expected_resampled_ts",
     (
@@ -77,4 +78,29 @@ def test_fit_transform_with_nans(daily_exog_ts_diff_endings):
     resampler = ResampleWithDistributionTransform(
         in_column="regressor_exog", inplace=True, distribution_column="target"
     )
-    daily_exog_ts_diff_endings.fit_transform([resampler])
+    _ = resampler.fit_transform(daily_exog_ts_diff_endings)
+
+
+@pytest.mark.filterwarnings("ignore: Regressors info might be incorrect.")
+@pytest.mark.parametrize(
+    "inplace, in_column_regressor, out_column, expected_regressors",
+    [
+        (True, False, None, []),
+        (False, False, "output_regressor", []),
+        (False, True, "output_regressor", ["output_regressor"]),
+    ],
+)
+def test_get_regressors_info(
+    daily_exog_ts, inplace, in_column_regressor, out_column, expected_regressors, in_column="regressor_exog"
+):
+    daily_exog_ts = daily_exog_ts["ts"]
+    if in_column_regressor:
+        daily_exog_ts._regressors.append(in_column)
+    else:
+        daily_exog_ts._regressors.remove(in_column)
+    resampler = ResampleWithDistributionTransform(
+        in_column=in_column, inplace=inplace, distribution_column="target", out_column=out_column
+    )
+    resampler.fit(daily_exog_ts)
+    regressors_info = resampler.get_regressors_info()
+    assert sorted(regressors_info) == sorted(expected_regressors)