Merge pull request #11 from yu9824/dev

v2.1.1 release

Fix #10 .

README.md

@@ -119,7 +119,9 @@ from sklearn.model_selection import cross_validate
 kf = KFold(n_splits=5, shuffle=True, random_state=334)
 print(cross_validate(estimator, X, y, cv=kf))
 ```
+
 OR
+
 ```python
 from sklearn.model_selection import cross_validate
 
@@ -136,10 +138,20 @@ If you want to run the notebook in example directory, you will need to additiona
 
 ## Parallelization (since v2.1.0)
 
-This algorithm is very computationally intensive and takes a lot of computation time.
-To solve this problem, we have implemented parallelization and optimized the algorithm since v2.1.0.
+This algorithm is very computationally intensive and takes a lot of time.
+To solve this problem, I have implemented parallelization and optimized the algorithm since v2.1.0.
 `n_jobs` can be specified for parallelization as in the scikit-learn-like api.
 
+```python
+# parallelization KFold
+kf = KFold(n_splits=5, n_jobs=-1)
+
+# parallelization train_test_split
+X_train, X_test, y_train, y_test = train_test_split(
+    X, y, test_size=0.2, n_jobs=-1
+)
+```
+
 
 ## LICENSE
 
@@ -176,5 +188,9 @@ Copyright (c) 2021 yu9824
 - Optimize algorithm
 - Deal with Large number of data.
   - parallel calculation when calculating distance (Add `n_jobs` argument)
-  - recursion number settings
+  - replacing recursive functions with for-loops
 - Add other than "euclidean" calculation methods (Add `metric` argument)
+
+### v2.1.1
+
+- Fix bug when `metric="nan_euclidean"`.

kennard_stone/__init__.py

@@ -1,6 +1,6 @@
 from .kennard_stone import KFold, train_test_split
 
-__version__ = "2.1.0"
+__version__ = "2.1.1"
 __license__ = "MIT"
 __author__ = "yu9824"
 __copyright__ = "Copyright © 2021 yu9824"

kennard_stone/kennard_stone.py

@@ -4,13 +4,13 @@
 
 from typing import overload, Union, Optional, Generator
 
-# deprecated in Python >= 3.9
+# The fllowing has deprecated in Python >= 3.9
 from typing import List, Set
+
 from itertools import chain
 import warnings
 
 import numpy as np
-
 from sklearn.model_selection._split import BaseShuffleSplit
 from sklearn.model_selection._split import _BaseKFold
 from sklearn.model_selection._split import _validate_shuffle_split
@@ -44,7 +44,8 @@ def __init__(
         *,
         metric: str = "euclidean",
         n_jobs: Optional[int] = None,
-        **kwargs,
+        random_state: None = None,
+        shuffle: None = None,
     ) -> None:
         """K-Folds cross-validator using the Kennard-Stone algorithm.
 
@@ -78,15 +79,15 @@ def __init__(
         self.metric = metric
         self.n_jobs = n_jobs
 
-        if "shuffle" in kwargs:
+        if shuffle is not None:
             warnings.warn(
                 "`shuffle` is unnecessary because it is always shuffled"
                 " in this algorithm.",
                 UserWarning,
             )
         del self.shuffle
 
-        if "random_state" in kwargs:
+        if random_state is not None:
             warnings.warn(
                 "`random_state` is unnecessary since it is uniquely determined"
                 " in this algorithm.",
@@ -166,7 +167,8 @@ def train_test_split(
     train_size: Optional[Union[float, int]] = None,
     metric: str = "euclidean",
     n_jobs: Optional[int] = None,
-    **kwargs,
+    random_state: None = None,
+    shuffle: None = None,
 ) -> list:
     """Split arrays or matrices into train and test subsets using the
     Kennard-Stone algorithm.
@@ -223,14 +225,14 @@ def train_test_split(
     ------
     ValueError
     """
-    if "shuffle" in kwargs:
+    if shuffle is not None:
         warnings.warn(
             "`shuffle` is unnecessary because it is always shuffled"
             " in this algorithm.",
             UserWarning,
         )
 
-    if "random_state" in kwargs:
+    if random_state is not None:
         warnings.warn(
             "`random_state` is unnecessary since it is uniquely determined"
             " in this algorithm.",
@@ -308,7 +310,14 @@ def __init__(
 
     def get_indexes(self, X) -> List[List[int]]:
         # check input array
-        X: np.ndarray = check_array(X, ensure_2d=True, dtype="numeric")
+        X: np.ndarray = check_array(
+            X,
+            ensure_2d=True,
+            dtype="numeric",
+            force_all_finite="allow-nan"
+            if self.metric == "nan_euclidean"
+            else True,
+        )
         n_samples = X.shape[0]
 
         # drop no variance
@@ -343,8 +352,6 @@ def get_indexes(self, X) -> List[List[int]]:
 
         distance_min = self.distance_matrix[idx_farthest, :]
 
-        # recursion limit settings
-
         # params
         indexes_selected = idx_farthest
         lst_indexes_selected_prev = [[] for _ in range(self.n_groups)]

tests/test_nan_euclidean.py

@@ -0,0 +1,12 @@
+from sklearn.datasets import load_diabetes
+
+from kennard_stone.kennard_stone import _KennardStone
+
+
+def test_nan_euclidean():
+    X = load_diabetes(as_frame=True).data.copy()
+
+    X.iloc[1, 1] = float("nan")
+
+    ks = _KennardStone(n_groups=1, metric="nan_euclidean")
+    ks.get_indexes(X)