diff --git a/src/timeseriesflattenerv2/feature_specs.py b/src/timeseriesflattenerv2/feature_specs.py
index dad1737f..0192f0c6 100644
--- a/src/timeseriesflattenerv2/feature_specs.py
+++ b/src/timeseriesflattenerv2/feature_specs.py
@@ -2,6 +2,7 @@
 from dataclasses import dataclass
 from typing import Protocol, Sequence, Union
 
+import pandas as pd
 import polars as pl
 
 ValueType = Union[int, float, str]
@@ -34,15 +35,23 @@ def to_lazyframe_with_uuid(self) -> pl.LazyFrame:
         return self.df
 
 
-@dataclass(frozen=True)
+@dataclass
 class ValueFrame:
     """A frame that contains the values of a time series."""
 
-    df: pl.LazyFrame
+    df: pl.LazyFrame | pd.DataFrame
     value_type: str
     entity_id_col_name: str = default_entity_id_col_name
     value_timestamp_col_name: str = "value_timestamp"
 
+    @property
+    def lazyframe(self) -> pl.LazyFrame:
+        return self.df if isinstance(self.df, pl.LazyFrame) else pl.from_pandas(self.df).lazy()
+
+    @property
+    def eagerframe(self) -> pl.DataFrame:
+        return self.df.collect() if isinstance(self.df, pl.LazyFrame) else pl.from_pandas(self.df)
+
 
 @dataclass(frozen=True)
 class SlicedFrame:
diff --git a/src/timeseriesflattenerv2/flattener.py b/src/timeseriesflattenerv2/flattener.py
index fa8f7e5a..48961eac 100644
--- a/src/timeseriesflattenerv2/flattener.py
+++ b/src/timeseriesflattenerv2/flattener.py
@@ -79,7 +79,7 @@ def _get_timedelta_frame(
 ) -> TimedeltaFrame:
     # Join the prediction time dataframe
     joined_frame = predictiontime_frame.to_lazyframe_with_uuid().join(
-        value_frame.df, on=predictiontime_frame.entity_id_col_name
+        value_frame.lazyframe, on=predictiontime_frame.entity_id_col_name
     )
 
     # Get timedelta
@@ -134,7 +134,7 @@ def aggregate_timeseries(self, specs: Sequence[ValueSpecification]) -> Aggregate
                     predictiontime_frame=self.predictiontime_frame, spec=spec
                 )
             )
-            .map(lambda x: x.df)
+            .map(lambda x: x.lazyframe)
             .to_list()
         )
         return AggregatedValueFrame(df=_horizontally_concatenate_dfs(dfs))
diff --git a/src/timeseriesflattenerv2/test_flattener.py b/src/timeseriesflattenerv2/test_flattener.py
index 2057694b..8e752543 100644
--- a/src/timeseriesflattenerv2/test_flattener.py
+++ b/src/timeseriesflattenerv2/test_flattener.py
@@ -17,7 +17,9 @@
 
 
 def assert_frame_equal(left: pl.DataFrame, right: pl.DataFrame):
-    polars_testing.assert_frame_equal(left, right, check_dtype=False, check_column_order=False)
+    polars_testing.assert_frame_equal(
+        left, right, check_dtype=False, check_column_order=False, check_row_order=False
+    )
 
 
 def test_flattener():