feat: Add ml.metrics.pairwise.euclidean_distance (#397)

Author: ashleyxuu

bigframes/ml/metrics/pairwise.py

@@ -50,3 +50,19 @@ def paired_manhattan_distance(
 paired_manhattan_distance.__doc__ = inspect.getdoc(
  vendored_metrics_pairwise.paired_manhattan_distance
 )
+
+
+def paired_euclidean_distances(
+ X: Union[bpd.DataFrame, bpd.Series], Y: Union[bpd.DataFrame, bpd.Series]
+) -> bpd.DataFrame:
+ X, Y = utils.convert_to_dataframe(X, Y)
+ if len(X.columns) != 1 or len(Y.columns) != 1:
+ raise ValueError("Inputs X and Y can only contain 1 column.")
+
+ base_bqml = core.BaseBqml(session=X._session)
+ return base_bqml.distance(X, Y, type="EUCLIDEAN", name="euclidean_distance")
+
+
+paired_euclidean_distances.__doc__ = inspect.getdoc(
+ vendored_metrics_pairwise.paired_euclidean_distances
+)

tests/system/small/ml/test_metrics_pairwise.py

@@ -47,3 +47,19 @@ def test_paired_manhattan_distance():
  pd.testing.assert_frame_equal(
  result.to_pandas(), expected_pd_df, check_dtype=False, check_index_type=False
  )
+
+
+def test_paired_euclidean_distances():
+ x_col = [np.array([4.1, 0.5, 1.0])]
+ y_col = [np.array([3.0, 0.0, 2.5])]
+ X = bpd.read_pandas(pd.DataFrame({"X": x_col}))
+ Y = bpd.read_pandas(pd.DataFrame({"Y": y_col}))
+
+ result = metrics.pairwise.paired_euclidean_distances(X, Y)
+ expected_pd_df = pd.DataFrame(
+ {"X": x_col, "Y": y_col, "euclidean_distance": [1.926136]}
+ )
+
+ pd.testing.assert_frame_equal(
+ result.to_pandas(), expected_pd_df, check_dtype=False, check_index_type=False
+ )

third_party/bigframes_vendored/sklearn/metrics/pairwise.py

@@ -39,3 +39,18 @@ def paired_manhattan_distance(X, Y) -> bpd.DataFrame:
  bigframes.dataframe.DataFrame: DataFrame with columns of X, Y and manhattan_distance
  """
  raise NotImplementedError(constants.ABSTRACT_METHOD_ERROR_MESSAGE)
+
+
+def paired_euclidean_distances(X, Y) -> bpd.DataFrame:
+ """Compute the paired euclidean distances between X and Y.
+
+ Args:
+ X (Series or single column DataFrame of array of numeric type):
+ Input data.
+ Y (Series or single column DataFrame of array of numeric type):
+ Input data. X and Y are mapped by indexes, must have the same index.
+
+ Returns:
+ bigframes.dataframe.DataFrame: DataFrame with columns of X, Y and euclidean_distance
+ """
+ raise NotImplementedError(constants.ABSTRACT_METHOD_ERROR_MESSAGE)