Categorical encoders

CHANGELOG.md

@@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Added
 - Method TSDataset.info ([#409](https://github.com/tinkoff-ai/etna/pull/409))
 - DifferencingTransform ([#414](https://github.com/tinkoff-ai/etna/pull/414))
+- OneHotEncoderTransform and LabelEncoderTransform ([#431](https://github.com/tinkoff-ai/etna/pull/431))
 
 ## Changed
 - Change method TSDataset.describe ([#409](https://github.com/tinkoff-ai/etna/pull/409))

etna/transforms/__init__.py

@@ -6,7 +6,9 @@
 from etna.transforms.decomposition import STLTransform
 from etna.transforms.decomposition import TheilSenTrendTransform
 from etna.transforms.decomposition import TrendTransform
+from etna.transforms.encoders import LabelEncoderTransform
 from etna.transforms.encoders import MeanSegmentEncoderTransform
+from etna.transforms.encoders import OneHotEncoderTransform
 from etna.transforms.encoders import SegmentEncoderTransform
 from etna.transforms.feature_selection import FilterFeaturesTransform
 from etna.transforms.feature_selection import GaleShapleyFeatureSelectionTransform

etna/transforms/encoders/__init__.py

@@ -1,2 +1,4 @@
+from etna.transforms.encoders.categorical import LabelEncoderTransform
+from etna.transforms.encoders.categorical import OneHotEncoderTransform
 from etna.transforms.encoders.mean_segment_encoder import MeanSegmentEncoderTransform
 from etna.transforms.encoders.segment_encoder import SegmentEncoderTransform

etna/transforms/encoders/categorical.py

@@ -0,0 +1,235 @@
+import warnings
+from enum import Enum
+from typing import Optional
+
+import numpy as np
+import pandas as pd
+from sklearn import preprocessing
+from sklearn.utils._encode import _check_unknown
+from sklearn.utils._encode import _encode
+
+from etna.transforms.base import PerSegmentWrapper
+from etna.transforms.base import Transform
+
+
+class ImputerMode(str, Enum):
+    """Enum for different imputation strategy."""
+
+    new_value = "new_value"
+    mean = "mean"
+    none = "none"
+
+
+class _LabelEncoder(preprocessing.LabelEncoder):
+    def transform(self, y: pd.Series, strategy: str):
+        diff = _check_unknown(y, known_values=self.classes_)
+
+        index = np.where(np.isin(y, diff))[0]
+
+        encoded = _encode(y, uniques=self.classes_, check_unknown=False).astype(float)
+
+        if strategy == "None":
+            filling_value = None
+        elif strategy == "new_value":
+            filling_value = -1
+        elif strategy == "mean":
+            filling_value = np.mean(encoded[~np.isin(y, diff)])
+        else:
+            raise ValueError(f"The strategy '{strategy}' doesn't exist")
+
+        encoded[index] = filling_value
+        return encoded
+
+
+class _OneSegmentLabelEncoderTransform(Transform):
+    """Replace the values in the column with the Label encoding."""
+
+    def __init__(self, in_column: str, out_column: str, strategy: str, inplace: bool):
+        """
+        Create instance of _OneSegmentLabelEncoderTransform.
+
+        Parameters
+        ----------
+        in_column:
+            name of column to apply transform to
+        out_column:
+            name of added column.
+        strategy:
+            filling encoding in not fitted values:
+            - If "new_value", then replace missing dates with '-1'
+            - If "mean", then replace missing dates using the mean in encoded column
+            - If "none", then replace missing dates with None
+        inplace:
+            if True, apply transform inplace to in_column, if False, add transformed column to dataset
+        """
+        self.in_column = in_column
+        self.out_column = out_column
+        self.strategy = strategy
+        self.le = _LabelEncoder()
+        self.inplace = inplace
+
+    def _get_column_name(self) -> str:
+        """Get the `out_column` depending on the transform's parameters."""
+        if self.inplace and self.out_column:
+            warnings.warn("Transformation will be applied inplace, out_column param will be ignored")
+        if self.inplace:
+            return self.in_column
+        if self.out_column:
+            return self.out_column
+        if self.in_column.startswith("regressor"):
+            temp_transform = LabelEncoderTransform(
+                in_column=self.in_column, inplace=self.inplace, out_column=self.out_column, strategy=self.strategy
+            )
+            return f"regressor_{temp_transform.__repr__()}"
+        temp_transform = LabelEncoderTransform(
+            in_column=self.in_column, inplace=self.inplace, out_column=self.out_column, strategy=self.strategy
+        )
+        return temp_transform.__repr__()
+
+    def fit(self, df: pd.DataFrame) -> "_OneSegmentLabelEncoderTransform":
+        """
+        Fit Label encoder.
+
+        Parameters
+        ----------
+        df:
+            dataframe with data to fit the transform.
+        Returns
+        -------
+        self
+        """
+        self.le.fit(df[self.in_column])
+        return self
+
+    def transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """
+        Encode the `in_column` by fitted Label encoder.
+
+        Parameters
+        ----------
+        df
+            dataframe with data to transform.
+        Returns
+        -------
+        result dataframe
+        """
+        result_df = df.copy()
+        result_df[self._get_column_name()] = self.le.transform(df[self.in_column], self.strategy)
+        return result_df
+
+
+class LabelEncoderTransform(PerSegmentWrapper):
+    """Encode categorical feature with value between 0 and n_classes-1."""
+
+    def __init__(
+        self, in_column: str, inplace: bool = True, out_column: Optional[str] = None, strategy: str = ImputerMode.mean
+    ):
+        """
+        Init LabelEncoderTransform.
+
+        Parameters
+        ----------
+        in_column:
+            name of column to be transformed
+        inplace:
+            if True, apply transform inplace to in_column, if False, add transformed column to dataset
+        out_column:
+            name of added column. If not given, use `self.__repr__()` or `regressor_{self.__repr__()}` if it is a regressor
+        strategy:
+            filling encoding in not fitted values:
+            - If "new_value", then replace missing values with '-1'
+            - If "mean", then replace missing values using the mean in encoded column
+            - If "none", then replace missing values with None
+        """
+        self.in_column = in_column
+        self.inplace = inplace
+        self.strategy = strategy
+        self.out_column = out_column
+        super().__init__(
+            transform=_OneSegmentLabelEncoderTransform(
+                in_column=self.in_column, out_column=self.out_column, strategy=self.strategy, inplace=self.inplace
+            )
+        )
+
+
+class _OneSegmentOneHotEncoderTransform(Transform):
+    """Create one-hot encoding columns."""
+
+    def __init__(self, in_column: str, out_column: str):
+        """
+        Create instance of _OneSegmentOneHotEncoderTransform.
+
+        Parameters
+        ----------
+        in_column:
+            name of column to apply transform to
+        out_column:
+            name of added column
+        """
+        self.in_column = in_column
+        self.out_column = out_column
+        self.ohe = preprocessing.OneHotEncoder(handle_unknown="ignore", sparse=False)
+
+    def _get_column_name(self) -> str:
+        """Get the `out_column` depending on the transform's parameters."""
+        if self.out_column:
+            return self.out_column
+        if self.in_column.startswith("regressor"):
+            temp_transform = OneHotEncoderTransform(in_column=self.in_column, out_column=self.out_column)
+            return f"regressor_{temp_transform.__repr__()}"
+        temp_transform = OneHotEncoderTransform(in_column=self.in_column, out_column=self.out_column)
+        return temp_transform.__repr__()
+
+    def fit(self, df: pd.DataFrame) -> "_OneSegmentOneHotEncoderTransform":
+        """
+        Fit One Hot encoder.
+
+        Parameters
+        ----------
+        df:
+            dataframe with data to fit the transform.
+        Returns
+        -------
+        self
+        """
+        self.ohe.fit(np.array(df[self.in_column]).reshape(-1, 1))
+        return self
+
+    def transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        """
+        Encode the `in_column` by fitted Label Binarize encoder.
+
+        Parameters
+        ----------
+        df
+            dataframe with data to transform.
+        Returns
+        -------
+        result dataframe
+        """
+        result_df = df.copy()
+        result_df[
+            [self._get_column_name() + "_" + str(i) for i in range(len(self.ohe.categories_[0]))]
+        ] = self.ohe.transform(np.array(df[self.in_column]).reshape(-1, 1))
+        return result_df
+
+
+class OneHotEncoderTransform(PerSegmentWrapper):
+    """Encode categorical feature as a one-hot numeric features."""
+
+    def __init__(self, in_column: str, out_column: Optional[str] = None):
+        """
+        Init OneHotEncoderTransform.
+
+        Parameters
+        ----------
+        in_column:
+            name of column to be encoded
+        out_column:
+            prefix of names of added columns. If not given, use `self.__repr__()` or `regressor_{self.__repr__()}` if it is a regressor
+        """
+        self.in_column = in_column
+        self.out_column = out_column
+        super().__init__(
+            transform=_OneSegmentOneHotEncoderTransform(in_column=self.in_column, out_column=self.out_column)
+        )

tests/test_transforms/test_encoders/test_categorical_transform.py

@@ -0,0 +1,93 @@
+import numpy as np
+import pytest
+
+from etna.datasets import TSDataset
+from etna.datasets import generate_periodic_df
+from etna.transforms.encoders.categorical import LabelEncoderTransform
+from etna.transforms.encoders.categorical import OneHotEncoderTransform
+
+
+@pytest.fixture
+def two_df_with_new_values():
+    df1 = TSDataset.to_dataset(
+        generate_periodic_df(periods=3, start_time="2020-01-01", scale=10, period=2, n_segments=2)
+    )
+    df2 = TSDataset.to_dataset(
+        generate_periodic_df(periods=3, start_time="2020-01-01", scale=10, period=3, n_segments=2)
+    )
+    return df1, df2
+
+
+@pytest.fixture
+def df_for_categorical_encoding():
+    return TSDataset.to_dataset(
+        generate_periodic_df(periods=4, start_time="2020-01-01", scale=10, period=3, n_segments=2)
+    )
+
+
+def test_label_encoder(df_for_categorical_encoding):
+    """Test LabelEncoderTransform correct works."""
+    le = LabelEncoderTransform(in_column="target")
+    le.fit(df_for_categorical_encoding)
+    expected_values = np.array([[0, 1], [1, 0], [2, 0], [0, 1]])
+    np.testing.assert_array_almost_equal(le.transform(df_for_categorical_encoding).values, expected_values)
+
+
+@pytest.mark.parametrize(
+    "strategy, expected_values",
+    [
+        ("new_value", np.array([[0, 0], [1, -1], [-1, -1]])),
+        ("None", np.array([[0, 0], [1, np.nan], [np.nan, np.nan]])),
+        ("mean", np.array([[0, 0], [1, 0], [0.5, 0]])),
+    ],
+)
+def test_new_value_label(two_df_with_new_values, strategy, expected_values):
+    """Test LabelEncoderTransform correct works with unknown values."""
+    df1, df2 = two_df_with_new_values
+    le = LabelEncoderTransform(in_column="target", strategy=strategy)
+    le.fit(df1)
+    np.testing.assert_array_almost_equal(le.transform(df2).values, expected_values)
+
+
+def test_value_error_label(df_for_categorical_encoding):
+    """Test LabelEncoderTransform with wrong strategy."""
+    with pytest.raises(ValueError, match="The strategy"):
+        le = LabelEncoderTransform(in_column="target", strategy="new_vlue")
+        le.fit(df_for_categorical_encoding)
+        le.transform(df_for_categorical_encoding)
+
+
+def test_ohe(df_for_categorical_encoding):
+    """Test OneHotEncoderTransform correct works."""
+    ohe = OneHotEncoderTransform(in_column="target")
+    ohe.fit(df_for_categorical_encoding)
+    expected_values = np.array(
+        [
+            [1.0, 0.0, 0.0, 5.0, 0.0, 1.0, 5.0],
+            [0.0, 1.0, 0.0, 8.0, 1.0, 0.0, 0.0],
+            [0.0, 0.0, 1.0, 9.0, 1.0, 0.0, 0.0],
+            [1.0, 0.0, 0.0, 5.0, 0.0, 1.0, 5.0],
+        ]
+    )
+    np.testing.assert_array_almost_equal(ohe.transform(df_for_categorical_encoding).values, expected_values)
+
+
+def test_new_value_ohe(two_df_with_new_values):
+    """Test OneHotEncoderTransform correct works with unknown values."""
+    expected_values = np.array(
+        [[5.0, 1.0, 0.0, 5.0, 1.0, 0.0], [8.0, 0.0, 1.0, 0.0, 0.0, 0.0], [9.0, 0.0, 0.0, 0.0, 0.0, 0.0]]
+    )
+    df1, df2 = two_df_with_new_values
+    ohe = OneHotEncoderTransform(in_column="target", out_column="targets")
+    ohe.fit(df1)
+    np.testing.assert_array_almost_equal(ohe.transform(df2).values, expected_values)
+
+
+def test_naming_ohe(two_df_with_new_values):
+    """Test OneHotEncoderTransform gives the correct columns."""
+    df1, df2 = two_df_with_new_values
+    ohe = OneHotEncoderTransform(in_column="target", out_column="targets")
+    ohe.fit(df1)
+    segments = ["segment_0", "segment_1"]
+    target = ["target", "targets_0", "targets_1"]
+    assert set([(i, j) for i in segments for j in target]) == set(ohe.transform(df2).columns.values)