feat(python): Add median stat to Series.describe (#8118)

pola-rs · Apr 10, 2023 · b2f7eb8 · b2f7eb8
1 parent 5ad0832
commit b2f7eb8
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 15 deletions.
diff --git a/py-polars/polars/series/series.py b/py-polars/polars/series/series.py
@@ -1157,18 +1157,19 @@ def describe(self) -> DataFrame:
         --------
         >>> series_num = pl.Series([1, 2, 3, 4, 5])
         >>> series_num.describe()
-        shape: (6, 2)
+        shape: (7, 2)
         ┌────────────┬──────────┐
         │ statistic  ┆ value    │
         │ ---        ┆ ---      │
         │ str        ┆ f64      │
         ╞════════════╪══════════╡
-        │ min        ┆ 1.0      │
-        │ max        ┆ 5.0      │
+        │ count      ┆ 5.0      │
         │ null_count ┆ 0.0      │
         │ mean       ┆ 3.0      │
         │ std        ┆ 1.581139 │
-        │ count      ┆ 5.0      │
+        │ min        ┆ 1.0      │
+        │ max        ┆ 5.0      │
+        │ median     ┆ 3.0      │
         └────────────┴──────────┘
 
         >>> series_str = pl.Series(["a", "a", None, "b", "c"])
@@ -1179,9 +1180,9 @@ def describe(self) -> DataFrame:
         │ ---        ┆ ---   │
         │ str        ┆ i64   │
         ╞════════════╪═══════╡
-        │ unique     ┆ 4     │
-        │ null_count ┆ 1     │
         │ count      ┆ 5     │
+        │ null_count ┆ 1     │
+        │ unique     ┆ 4     │
         └────────────┴───────┘
 
         """
@@ -1192,33 +1193,35 @@ def describe(self) -> DataFrame:
         elif self.is_numeric():
             s = self.cast(Float64)
             stats = {
-                "min": s.min(),
-                "max": s.max(),
+                "count": s.len(),
                 "null_count": s.null_count(),
                 "mean": s.mean(),
                 "std": s.std(),
-                "count": s.len(),
+                "min": s.min(),
+                "max": s.max(),
+                "median": s.median(),
             }
         elif self.is_boolean():
             stats = {
-                "sum": self.sum(),
-                "null_count": self.null_count(),
                 "count": self.len(),
+                "null_count": self.null_count(),
+                "sum": self.sum(),
             }
         elif self.is_utf8():
             stats = {
-                "unique": len(self.unique()),
-                "null_count": self.null_count(),
                 "count": self.len(),
+                "null_count": self.null_count(),
+                "unique": len(self.unique()),
             }
         elif self.is_temporal():
             # we coerce all to string, because a polars column
             # only has a single dtype and dates: datetime and count: int don't match
             stats = {
+                "count": str(self.len()),
+                "null_count": str(self.null_count()),
                 "min": str(self.dt.min()),
                 "max": str(self.dt.max()),
-                "null_count": str(self.null_count()),
-                "count": str(self.len()),
+                "median": str(self.dt.median()),
             }
         else:
             raise TypeError("This type is not supported")

diff --git a/py-polars/tests/unit/test_series.py b/py-polars/tests/unit/test_series.py
@@ -1130,6 +1130,7 @@ def test_describe() -> None:
         "min": 1.0,
         "null_count": 0.0,
         "std": 1.0,
+        "median": 2.0,
     }
     assert dict(float_s.describe().rows()) == {  # type: ignore[arg-type]
         "count": 3.0,
@@ -1138,6 +1139,7 @@ def test_describe() -> None:
         "min": 1.3,
         "null_count": 0.0,
         "std": 3.8109491381194442,
+        "median": 4.6,
     }
     assert dict(str_s.describe().rows()) == {  # type: ignore[arg-type]
         "count": 3,
@@ -1153,6 +1155,7 @@ def test_describe() -> None:
         "count": "3",
         "max": "2021-01-03",
         "min": "2021-01-01",
+        "median": "2021-01-02",
         "null_count": "0",
     }