feat: merge dev

Author: anhkhoangoho

qolmat/benchmark/comparator.py

@@ -168,13 +168,12 @@ def compare(
 
 
 class ComparatorBasedPattern(Comparator):
-
     def __init__(
         self,
         dict_models: Dict[str, Any],
         selected_columns: List[str],
         generator_holes: _HoleGenerator,
-         metrics: List = ["mae", "wmape", "KL"],
+        metrics: List = ["mae", "wmape", "KL"],
         search_params: Optional[Dict[str, Dict[str, Union[float, int, str]]]] = {},
         n_calls_opt: int = 10,
         num_patterns: int = 5,

qolmat/benchmark/metrics.py

@@ -133,6 +133,7 @@ def kl_divergence_1D(df1: pd.Series, df2: pd.Series) -> np.number:
     q = np.histogram(df2, bins=bins, density=True)[0]
     return scipy.stats.entropy(p + EPS, q + EPS)
 
+
 def kl_divergence_columnwise(
     df1: pd.DataFrame, df2: pd.DataFrame, df_mask: pd.DataFrame
 ) -> pd.Series:
@@ -230,7 +231,8 @@ def kolmogorov_smirnov_test(df1: pd.DataFrame, df2: pd.DataFrame) -> pd.Series:
 
 
 def total_variance_distance(df1: pd.DataFrame, df2: pd.DataFrame) -> pd.Series:
-    """Total variance distance for categorical features, based on TVComplement in https://github.com/sdv-dev/SDMetrics
+    """Total variance distance for categorical features
+    It is based on TVComplement in https://github.com/sdv-dev/SDMetrics
 
     Parameters
     ----------
@@ -436,7 +438,8 @@ def _sum_distance_col(col: pd.Series, col_size: int) -> pd.Series:
 
 
 def _sum_manhattan_distances(df1: pd.DataFrame) -> float:
-    """Sum Manhattan distances. It is based on https://www.geeksforgeeks.org/sum-manhattan-distances-pairs-points/
+    """Sum Manhattan distances.
+    It is based on https://www.geeksforgeeks.org/sum-manhattan-distances-pairs-points/
 
     Parameters
     ----------
@@ -451,7 +454,8 @@ def _sum_manhattan_distances(df1: pd.DataFrame) -> float:
 
 
 def sum_energy_distances(df1: pd.DataFrame, df2: pd.DataFrame) -> float:
-    """Sum of energy distances between df1 and df2. It is based on https://dcor.readthedocs.io/en/latest/theory.html#
+    """Sum of energy distances between df1 and df2.
+    It is based on https://dcor.readthedocs.io/en/latest/theory.html#
 
     Parameters
     ----------
@@ -508,6 +512,7 @@ def sum_pairwise_distances(df1: pd.DataFrame, df2: pd.DataFrame, metric: str = "
 # Dataframe-wise metris   #
 ###########################
 
+
 def frechet_distance(
     df1: pd.DataFrame,
     df2: pd.DataFrame,
@@ -568,4 +573,4 @@ def frechet_distance(
     if normalized:
         return pd.Series((frechet_dist / df_true.shape[0]), index=["All"])
     else:
-        return pd.Series(np.repeat(frechet_dist, len(df1.columns)))
+        return pd.Series(np.repeat(frechet_dist, len(df1.columns)))

qolmat/tests/benchmark/test_metrics.py

@@ -24,10 +24,10 @@
 df = pd.DataFrame([[1, 1, 1, 1], [1, 2, 2, 1], [2, 2, 2, 2]])
 
 
-@pytest.mark.parametrize("df", [df])
-def test_em_sampler_convert_numpy(df: NDArray) -> None:
-    """Test converge Numpy for Impute EM"""
-    assert type(em_sampler.EM()._convert_numpy(df)) == np.ndarray
+# @pytest.mark.parametrize("df", [df])
+# def test_em_sampler_convert_numpy(df: NDArray) -> None:
+#     """Test converge Numpy for Impute EM"""
+#     assert type(em_sampler.EM()._convert_numpy(df)) == np.ndarray
 
 
 imputations_var = [