upper bound get_lags_for_frequency by context_length

pts/model/deepar/deepar_estimator.py

@@ -91,7 +91,9 @@ class DeepAREstimator(PyTorchEstimator):
         )
         self.scaling = scaling
         self.lags_seq = (
-            lags_seq if lags_seq is not None else get_lags_for_frequency(freq_str=freq)
+            lags_seq
+            if lags_seq is not None
+            else get_lags_for_frequency(freq_str=freq, lag_ub=self.context_length)
         )
         self.time_features = (
             time_features

pts/model/simple_feedforward/simple_feedforward_estimator.py

@@ -9,11 +9,6 @@ from gluonts.torch.model.predictor import PyTorchPredictor
 from gluonts.torch.modules.distribution_output import DistributionOutput
 from gluonts.model.predictor import Predictor
 from gluonts.dataset.field_names import FieldName
-from gluonts.time_feature import (
-    TimeFeature,
-    get_lags_for_frequency,
-    time_features_from_frequency_str,
-)
 from gluonts.transform import (
     Transformation,
     Chain,