Change ProphetModel and SARIMAXModel according to latest architec…

…ture (#549)
tinkoff-ai · Feb 18, 2022 · 8f594cf · 8f594cf
1 parent 2242d40
commit 8f594cf
Show file tree

Hide file tree

Showing 11 changed files with 112 additions and 215 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -35,6 +35,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - 
 - Rename `_CatBoostModel`, `_HoltWintersModel`, `_SklearnModel` ([#543](https://github.com/tinkoff-ai/etna/pull/543))
 - 
+- Rename `_SARIMAXModel` and `_ProphetModel`, make `SARIMAXModel` and `ProphetModel` inherit from `PerSegmentPredictionIntervalModel` ([#549](https://github.com/tinkoff-ai/etna/pull/549))
+- 
 ### Fixed
 - Fix `TSDataset._update_regressors` logic removing the regressors ([#489](https://github.com/tinkoff-ai/etna/pull/489)) 
 - Fix `TSDataset.info`, `TSDataset.describe` methods ([#519](https://github.com/tinkoff-ai/etna/pull/519))

diff --git a/etna/models/base.py b/etna/models/base.py
@@ -10,6 +10,7 @@
 from typing import Sequence
 from typing import Union
 
+import numpy as np
 import pandas as pd
 
 from etna.core.mixins import BaseMixin
@@ -181,7 +182,6 @@ def __init__(self, base_model: Any):
             Internal model which will be used to forecast segments, expected to have fit/predict interface
         """
         self._base_model = base_model
-        self._segments: Optional[List[str]] = None
         self._models: Optional[Dict[str, Any]] = None
 
     @log_decorator
@@ -198,7 +198,6 @@ def fit(self, ts: TSDataset) -> "PerSegmentBaseModel":
         self:
             Model after fit
         """
-        self._segments = ts.segments
         self._models = {}
         for segment in ts.segments:
             self._models[segment] = deepcopy(self._base_model)
@@ -223,7 +222,7 @@ def get_model(self) -> Dict[str, Any]:
            dictionary where key is segment and value is internal model
         """
         if self._models is None:
-            raise ValueError("Can not get the dict with base models from not fitted model!")
+            raise ValueError("Can not get the dict with base models, the model is not fitted!")
         return self._models
 
     @staticmethod
@@ -235,17 +234,12 @@ def _forecast_segment(model: Any, segment: str, ts: TSDataset, *args, **kwargs)
         dates = segment_features["timestamp"]
         dates.reset_index(drop=True, inplace=True)
         segment_predict = model.predict(df=segment_features, *args, **kwargs)
-        segment_predict = pd.DataFrame({"target": segment_predict})
+        if isinstance(segment_predict, np.ndarray):
+            segment_predict = pd.DataFrame({"target": segment_predict})
         segment_predict["segment"] = segment
         segment_predict["timestamp"] = dates
         return segment_predict
 
-    def _build_models(self):
-        """Create a dict with models for each segment (if required)."""
-        self._models = {}
-        for segment in self._segments:  # type: ignore
-            self._models[segment] = deepcopy(self._base_model)
-
 
 class PerSegmentModel(PerSegmentBaseModel, ForecastAbstractModel):
     """Class for holding specific models for per-segment prediction."""
@@ -305,7 +299,7 @@ def __init__(self, base_model: Any):
         """
         super().__init__(base_model=base_model)
 
-    @abstractmethod
+    @log_decorator
     def forecast(
         self, ts: TSDataset, prediction_interval: bool = False, quantiles: Sequence[float] = (0.025, 0.975)
     ) -> TSDataset:

diff --git a/etna/models/catboost.py b/etna/models/catboost.py
@@ -36,6 +36,20 @@ def __init__(
         self._categorical = None
 
     def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_CatBoostAdapter":
+        """
+        Fit Catboost model.
+
+        Parameters
+        ----------
+        df:
+            Features dataframe
+        regressors:
+            List of the columns with regressors(ignored in this model)
+        Returns
+        -------
+        self:
+            Fitted model
+        """
         features = df.drop(columns=["timestamp", "target"])
         target = df["target"]
         self._categorical = features.select_dtypes(include=["category"]).columns.to_list()
@@ -44,6 +58,19 @@ def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_CatBoostAdapter":
         return self
 
     def predict(self, df: pd.DataFrame) -> np.ndarray:
+        """
+        Compute predictions from a Catboost model.
+
+        Parameters
+        ----------
+        df:
+            Features dataframe
+
+        Returns
+        -------
+        y_pred:
+            Array with predictions
+        """
         features = df.drop(columns=["timestamp", "target"])
         predict_pool = Pool(features, cat_features=self._categorical)
         pred = self.model.predict(predict_pool)

diff --git a/etna/models/holt_winters.py b/etna/models/holt_winters.py
@@ -7,6 +7,7 @@
 from typing import Tuple
 from typing import Union
 
+import numpy as np
 import pandas as pd
 from statsmodels.tsa.holtwinters import ExponentialSmoothing
 from statsmodels.tsa.holtwinters import HoltWintersResults
@@ -171,17 +172,18 @@ def __init__(
 
     def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_HoltWintersAdapter":
         """
-        Fits a Holt-Winters' model.
+        Fit Holt-Winters' model.
 
         Parameters
         ----------
         df:
             Features dataframe
-
+        regressors:
+            List of the columns with regressors(ignored in this model)
         Returns
         -------
-        self: _HoltWintersAdapter
-            fitted model
+        self:
+            Fitted model
         """
         self._check_df(df)
 
@@ -213,7 +215,7 @@ def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_HoltWintersAdapter":
         )
         return self
 
-    def predict(self, df: pd.DataFrame) -> pd.Series:
+    def predict(self, df: pd.DataFrame) -> np.ndarray:
         """
         Compute predictions from a Holt-Winters' model.
 
@@ -224,15 +226,15 @@ def predict(self, df: pd.DataFrame) -> pd.Series:
 
         Returns
         -------
-        y_pred: pd.Series
-            Series with predictions
+        y_pred:
+            Array with predictions
         """
         if self._result is None or self._model is None:
             raise ValueError("This model is not fitted! Fit the model before calling predict method!")
         self._check_df(df)
 
         forecast = self._result.predict(start=df["timestamp"].min(), end=df["timestamp"].max())
-        y_pred = pd.Series(data=forecast.values, name="target")
+        y_pred = forecast.values
         return y_pred
 
     def _check_df(self, df: pd.DataFrame):

diff --git a/etna/models/prophet.py b/etna/models/prophet.py
@@ -9,15 +9,13 @@
 import pandas as pd
 
 from etna import SETTINGS
-from etna.datasets import TSDataset
-from etna.models.base import PerSegmentModel
-from etna.models.base import log_decorator
+from etna.models.base import PerSegmentPredictionIntervalModel
 
 if SETTINGS.prophet_required:
     from prophet import Prophet
 
 
-class _ProphetModel:
+class _ProphetAdapter:
     """Class for holding Prophet model."""
 
     def __init__(
@@ -83,7 +81,7 @@ def __init__(
 
         self.regressor_columns: Optional[List[str]] = None
 
-    def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_ProphetModel":
+    def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_ProphetAdapter":
         """
         Fits a Prophet model.
 
@@ -104,9 +102,9 @@ def fit(self, df: pd.DataFrame, regressors: List[str]) -> "_ProphetModel":
         self.model.fit(prophet_df)
         return self
 
-    def predict(self, df: pd.DataFrame, prediction_interval: bool, quantiles: Sequence[float]):
+    def predict(self, df: pd.DataFrame, prediction_interval: bool, quantiles: Sequence[float]) -> pd.DataFrame:
         """
-        Compute Prophet predictions.
+        Compute predictions from a Prophet model.
 
         Parameters
         ----------
@@ -119,7 +117,7 @@ def predict(self, df: pd.DataFrame, prediction_interval: bool, quantiles: Sequen
 
         Returns
         -------
-        y_pred: pd.DataFrame
+        y_pred:
             DataFrame with predictions
         """
         df = df.reset_index()
@@ -134,10 +132,14 @@ def predict(self, df: pd.DataFrame, prediction_interval: bool, quantiles: Sequen
             for quantile in quantiles:
                 percentile = quantile * 100
                 y_pred[f"yhat_{quantile:.4g}"] = self.model.percentile(sim_values["yhat"], percentile, axis=1)
+        rename_dict = {
+            column: column.replace("yhat", "target") for column in y_pred.columns if column.startswith("yhat")
+        }
+        y_pred = y_pred.rename(rename_dict, axis=1)
         return y_pred
 
 
-class ProphetModel(PerSegmentModel):
+class ProphetModel(PerSegmentPredictionIntervalModel):
     """Class for holding Prophet model.
 
     Examples
@@ -296,7 +298,7 @@ def __init__(
         self.additional_seasonality_params = additional_seasonality_params
 
         super(ProphetModel, self).__init__(
-            base_model=_ProphetModel(
+            base_model=_ProphetAdapter(
                 growth=self.growth,
                 n_changepoints=self.n_changepoints,
                 changepoints=self.changepoints,
@@ -316,84 +318,3 @@ def __init__(
                 additional_seasonality_params=self.additional_seasonality_params,
             )
         )
-
-    @log_decorator
-    def fit(self, ts: TSDataset) -> "ProphetModel":
-        """Fit model."""
-        self._segments = ts.segments
-        self._build_models()
-
-        for segment in self._segments:
-            model = self._models[segment]  # type: ignore
-            segment_features = ts[:, segment, :]
-            segment_features = segment_features.dropna()
-            segment_features = segment_features.droplevel("segment", axis=1)
-            segment_features = segment_features.reset_index()
-            model.fit(df=segment_features, regressors=ts.regressors)
-        return self
-
-    @staticmethod
-    def _forecast_one_segment(
-        model,
-        segment: Union[str, List[str]],
-        ts: TSDataset,
-        prediction_interval: bool,
-        quantiles: Sequence[float],
-    ) -> pd.DataFrame:
-        segment_features = ts[:, segment, :]
-        segment_features = segment_features.droplevel("segment", axis=1)
-        segment_features = segment_features.reset_index()
-        dates = segment_features["timestamp"]
-        dates.reset_index(drop=True, inplace=True)
-        segment_predict = model.predict(
-            df=segment_features, prediction_interval=prediction_interval, quantiles=quantiles
-        )
-        rename_dict = {
-            column: column.replace("yhat", "target") for column in segment_predict.columns if column.startswith("yhat")
-        }
-        segment_predict = segment_predict.rename(rename_dict, axis=1)
-        segment_predict["segment"] = segment
-        segment_predict["timestamp"] = dates
-        return segment_predict
-
-    @log_decorator
-    def forecast(
-        self, ts: TSDataset, prediction_interval: bool = False, quantiles: Sequence[float] = (0.025, 0.975)
-    ) -> TSDataset:
-        """Make predictions.
-
-        Parameters
-        ----------
-        ts:
-            Dataframe with features
-        prediction_interval:
-            If True returns prediction interval for forecast
-        quantiles:
-            Levels of prediction distribution. By default 2.5% and 97.5% taken to form a 95% prediction interval
-
-        Returns
-        -------
-        TSDataset
-            Models result
-        """
-        if self._segments is None:
-            raise ValueError("The model is not fitted yet, use fit() to train it")
-
-        result_list = list()
-        for segment in self._segments:
-            model = self._models[segment]  # type: ignore
-
-            segment_predict = self._forecast_one_segment(model, segment, ts, prediction_interval, quantiles)
-            result_list.append(segment_predict)
-
-        # need real case to test
-        result_df = pd.concat(result_list, ignore_index=True)
-        result_df = result_df.set_index(["timestamp", "segment"])
-        df = ts.to_pandas(flatten=True)
-        df = df.set_index(["timestamp", "segment"])
-        df = df.combine_first(result_df).reset_index()
-
-        df = TSDataset.to_dataset(df)
-        ts.df = df
-        ts.inverse_transform()
-        return ts