Remove parameters freq and dates from statsmodels-base models (#590)

* remove freq and dates attributes * lints * update changelog * lints * return filterwarning to script --------- Co-authored-by: Egor Baturin <[email protected]>
etna-team · Jan 31, 2025 · 8031c8c · 8031c8c
1 parent 064b60f
commit 8031c8c
Show file tree

Hide file tree

Showing 4 changed files with 1 addition and 54 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -47,7 +47,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Removed
 - **Breaking:** Remove `FutureMixin`, `OutliersTransform.outliers_timestamps` and `OutliersTransform.original_values` ([#577](https://github.com/etna-team/etna/pull/577))
--
+- **Breaking:** Remove parameters `freq` and `dates` from `SARIMAXModel` and `HoltWintersModel` ([#590](https://github.com/etna-team/etna/pull/590))
 - **Breaking:** Remove `pytorch-forecasting` models and utilities (`DeepARModel`, `TFTModel` and `PytorchForecastingDatasetBuilder`) ([#578](https://github.com/etna-team/etna/pull/578))
 -
 -

diff --git a/etna/models/holt_winters.py b/etna/models/holt_winters.py
@@ -1,5 +1,4 @@
 import warnings
-from datetime import datetime
 from typing import Dict
 from typing import List
 from typing import Optional
@@ -47,8 +46,6 @@ def __init__(
         initial_seasonal: Optional[Sequence[float]] = None,
         use_boxcox: Union[bool, str, float] = False,
         bounds: Optional[Dict[str, Tuple[float, float]]] = None,
-        dates: Optional[Sequence[datetime]] = None,
-        freq: Optional[str] = None,
         missing: str = "none",
         smoothing_level: Optional[float] = None,
         smoothing_trend: Optional[float] = None,
@@ -152,12 +149,6 @@ def __init__(
             for j=0,...,m-1 where m is the number of period in a full season.
             Use None to indicate a non-binding constraint, e.g., (0, None)
             constrains a parameter to be non-negative.
-        dates:
-            An array-like object of datetime objects. If a Pandas object is given
-            for endog, it is assumed to have a DateIndex.
-        freq:
-            The frequency of the time-series. A Pandas offset or 'B', 'D', 'W',
-            'M', 'A', or 'Q'. This is optional if dates are given.
         missing:
             Available options are 'none', 'drop', and 'raise'. If 'none', no nan
             checking is done. If 'drop', any observations with nans are dropped.
@@ -187,8 +178,6 @@ def __init__(
         self.initial_seasonal = initial_seasonal
         self.use_boxcox = use_boxcox
         self.bounds = bounds
-        self.dates = dates
-        self.freq = freq
         self.missing = missing
         self.smoothing_level = smoothing_level
         self.smoothing_trend = smoothing_trend
@@ -251,8 +240,6 @@ def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_HoltWintersAdapter":
             initial_seasonal=self.initial_seasonal,
             use_boxcox=self.use_boxcox,
             bounds=self.bounds,
-            dates=self.dates,
-            freq=self.freq,
             missing=self.missing,
         )
         self._result = self._model.fit(
@@ -495,8 +482,6 @@ def __init__(
         initial_seasonal: Optional[Sequence[float]] = None,
         use_boxcox: Union[bool, str, float] = False,
         bounds: Optional[Dict[str, Tuple[float, float]]] = None,
-        dates: Optional[Sequence[datetime]] = None,
-        freq: Optional[str] = None,
         missing: str = "none",
         smoothing_level: Optional[float] = None,
         smoothing_trend: Optional[float] = None,
@@ -600,12 +585,6 @@ def __init__(
             for j=0,...,m-1 where m is the number of period in a full season.
             Use None to indicate a non-binding constraint, e.g., (0, None)
             constrains a parameter to be non-negative.
-        dates:
-            An array-like object of datetime objects. If a Pandas object is given
-            for endog, it is assumed to have a DateIndex.
-        freq:
-            The frequency of the time-series. A Pandas offset or 'B', 'D', 'W',
-            'M', 'A', or 'Q'. This is optional if dates are given.
         missing:
             Available options are 'none', 'drop', and 'raise'. If 'none', no nan
             checking is done. If 'drop', any observations with nans are dropped.
@@ -635,8 +614,6 @@ def __init__(
         self.initial_seasonal = initial_seasonal
         self.use_boxcox = use_boxcox
         self.bounds = bounds
-        self.dates = dates
-        self.freq = freq
         self.missing = missing
         self.smoothing_level = smoothing_level
         self.smoothing_trend = smoothing_trend
@@ -655,8 +632,6 @@ def __init__(
                 initial_seasonal=self.initial_seasonal,
                 use_boxcox=self.use_boxcox,
                 bounds=self.bounds,
-                dates=self.dates,
-                freq=self.freq,
                 missing=self.missing,
                 smoothing_level=self.smoothing_level,
                 smoothing_trend=self.smoothing_trend,

diff --git a/etna/models/sarimax.py b/etna/models/sarimax.py
@@ -1,6 +1,5 @@
 import warnings
 from abc import abstractmethod
-from datetime import datetime
 from typing import Any
 from typing import Dict
 from typing import List
@@ -456,8 +455,6 @@ def __init__(
         concentrate_scale: bool = False,
         trend_offset: float = 1,
         use_exact_diffuse: bool = False,
-        dates: Optional[List[datetime]] = None,
-        freq: Optional[str] = None,
         missing: str = "none",
         validate_specification: bool = True,
         fit_params: Optional[Dict[str, Any]] = None,
@@ -537,12 +534,6 @@ def __init__(
             Whether or not to use exact diffuse initialization for non-stationary
             states. Default is False (in which case approximate diffuse
             initialization is used).
-        dates:
-            If no index is given by `endog` or `exog`, an array-like object of
-            datetime objects can be provided.
-        freq:
-            If no index is given by `endog` or `exog`, the frequency of the
-            time-series may be specified here as a Pandas offset or offset string.
         missing:
             Available options are 'none', 'drop', and 'raise'. If 'none', no nan
             checking is done. If 'drop', any observations with nans are dropped.
@@ -568,8 +559,6 @@ def __init__(
         self.concentrate_scale = concentrate_scale
         self.trend_offset = trend_offset
         self.use_exact_diffuse = use_exact_diffuse
-        self.dates = dates
-        self.freq = freq
         self.missing = missing
         self.validate_specification = validate_specification
         self.fit_params = fit_params if fit_params else {}
@@ -595,8 +584,6 @@ def _get_fit_results(self, endog: pd.Series, exog: pd.DataFrame):
             concentrate_scale=self.concentrate_scale,
             trend_offset=self.trend_offset,
             use_exact_diffuse=self.use_exact_diffuse,
-            dates=self.dates,
-            freq=self.freq,
             missing=self.missing,
             validate_specification=self.validate_specification,
             **self.kwargs,
@@ -641,8 +628,6 @@ def __init__(
         concentrate_scale: bool = False,
         trend_offset: float = 1,
         use_exact_diffuse: bool = False,
-        dates: Optional[List[datetime]] = None,
-        freq: Optional[str] = None,
         missing: str = "none",
         validate_specification: bool = True,
         fit_params: Optional[Dict[str, Any]] = None,
@@ -722,12 +707,6 @@ def __init__(
             Whether or not to use exact diffuse initialization for non-stationary
             states. Default is False (in which case approximate diffuse
             initialization is used).
-        dates:
-            If no index is given by `endog` or `exog`, an array-like object of
-            datetime objects can be provided.
-        freq:
-            If no index is given by `endog` or `exog`, the frequency of the
-            time-series may be specified here as a Pandas offset or offset string.
         missing:
             Available options are 'none', 'drop', and 'raise'. If 'none', no nan
             checking is done. If 'drop', any observations with nans are dropped.
@@ -753,8 +732,6 @@ def __init__(
         self.concentrate_scale = concentrate_scale
         self.trend_offset = trend_offset
         self.use_exact_diffuse = use_exact_diffuse
-        self.dates = dates
-        self.freq = freq
         self.missing = missing
         self.validate_specification = validate_specification
         self.fit_params = fit_params if fit_params else {}
@@ -774,8 +751,6 @@ def __init__(
                 concentrate_scale=self.concentrate_scale,
                 trend_offset=self.trend_offset,
                 use_exact_diffuse=self.use_exact_diffuse,
-                dates=self.dates,
-                freq=self.freq,
                 missing=self.missing,
                 validate_specification=self.validate_specification,
                 fit_params=self.fit_params,

diff --git a/pyproject.toml b/pyproject.toml
@@ -263,11 +263,8 @@ filterwarnings = [
     "ignore: The behavior of Timestamp.utcfromtimestamp is deprecated", # prophet
     "ignore: Starting from v1.9.0, `tensorboardX` has been removed as a dependency",
     "ignore: the `interpolation=` argument to percentile was renamed to `method=`", # pandas: old version uses deprecated numpy parameter
-    "ignore: Trying to infer the `batch_size` from an ambiguous collection", # pytorch_forecasting
     "ignore: Implicitly cleaning up <TemporaryDirectory",
-    "ignore: Call to deprecated class DeepARModel.",
     "ignore: dropout option adds dropout after all but last recurrent layer",
-    "ignore: Call to deprecated class TFTModel.",
     "ignore: X does not have valid feature names, but IsolationForest was fitted with feature names" # fixed here https://github.com/scikit-learn/scikit-learn/issues/25844
 ]
 markers = [