Merge pull request #520 from Aarhus-Psychiatry-Research/docs/511/add_…

…docstrings_to_aggregators

docs(#511): add docstrings to aggregators

docs/tutorials/04_from_legacy.ipynb

@@ -116,9 +116,11 @@
    ],
    "source": [
     "from timeseriesflattener import PredictorSpec as Version2PredictorSpec\n",
+    "\n",
     "# IMPORT CHANGED BELOW #\n",
-    "# from timeseriesflattener.v1.feature_specs.single_specs import PredictorSpec as Version1PredictorSpec\n",
+    "# from timeseriesflattener.v1.feature_specs.single_specs import PredictorSpec as Version1PredictorSpec # noqa: ERA001\n",
     "from timeseriesflattener.feature_specs.from_legacy import PredictorGroupSpec\n",
+    "\n",
     "# IMPORT CHANGED ABOVE #\n",
     "\n",
     "new_specs = PredictorGroupSpec(\n",

src/timeseriesflattener/aggregators.py

@@ -19,27 +19,35 @@ def new_col_name(self, previous_col_name: str) -> str:
 
 
 class MinAggregator(Aggregator):
+    """Returns the minimum value in the look window."""
+
     name: str = "min"
 
     def __call__(self, column_name: str) -> pl.Expr:
         return pl.col(column_name).min().alias(self.new_col_name(column_name))
 
 
 class MaxAggregator(Aggregator):
+    """Returns the maximum value in the look window."""
+
     name: str = "max"
 
     def __call__(self, column_name: str) -> pl.Expr:
         return pl.col(column_name).max().alias(self.new_col_name(column_name))
 
 
 class MeanAggregator(Aggregator):
+    """Returns the mean value in the look window."""
+
     name: str = "mean"
 
     def __call__(self, column_name: str) -> pl.Expr:
         return pl.col(column_name).mean().alias(self.new_col_name(column_name))
 
 
 class CountAggregator(Aggregator):
+    """Returns the count of non-null values in the look window."""
+
     name: str = "count"
 
     def __call__(self, column_name: str) -> pl.Expr:
@@ -48,6 +56,8 @@ def __call__(self, column_name: str) -> pl.Expr:
 
 @dataclass(frozen=True)
 class EarliestAggregator(Aggregator):
+    """Returns the earliest value in the look window."""
+
     timestamp_col_name: str
     name: str = "earliest"
 
@@ -62,6 +72,8 @@ def __call__(self, column_name: str) -> pl.Expr:
 
 @dataclass(frozen=True)
 class LatestAggregator(Aggregator):
+    """Returns the latest value in the look window"""
+
     timestamp_col_name: str
     name: str = "latest"
 
@@ -75,13 +87,17 @@ def __call__(self, column_name: str) -> pl.Expr:
 
 
 class SumAggregator(Aggregator):
+    """Returns the sum of all values in the look window."""
+
     name: str = "sum"
 
     def __call__(self, column_name: str) -> pl.Expr:
         return pl.col(column_name).sum().alias(self.new_col_name(column_name))
 
 
 class VarianceAggregator(Aggregator):
+    """Returns the variance of the values in the look window"""
+
     name: str = "var"
 
     def __call__(self, column_name: str) -> pl.Expr:
@@ -106,6 +122,8 @@ def __call__(self, column_name: str) -> pl.Expr:
 
 @dataclass(frozen=True)
 class SlopeAggregator(Aggregator):
+    """Returns the slope (i.e. the correlation between the timestamp and the value) in the look window."""
+
     timestamp_col_name: str
     name: str = "slope"