Add checking that binary matrix is not suitable for hypershpere overlap (#280)

* Add checking that binary matrix is not suitable for hypershpere overlap

* Update testing

Author: FanwangM

selector/measures/diversity.py

@@ -403,6 +403,11 @@ def hypersphere_overlap_of_subset(x: np.ndarray, x_subset: np.array) -> float:
     Agrafiotis, D. K.. (1997) Stochastic Algorithms for Maximizing Molecular Diversity.
     Journal of Chemical Information and Computer Sciences 37, 841-851.
     """
+    # check if x is binary matrix or not
+    if np.array_equal(x, x.astype(bool)):
+        raise ValueError(
+            "Input matrix cannot be binary because the diversity measurement is designed for continuous orthogonal features."
+        )
 
     # Find the maximum and minimum over each feature across all molecules.
     max_x = np.max(x, axis=0)

selector/measures/tests/test_diversity.py

@@ -256,7 +256,7 @@ def test_hypersphere_normalization_error():
 
 def test_hypersphere_radius_warning():
     """Test the hypersphere overlap method gives warning when radius is too large."""
-    corner_pts = np.array([[0.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 1.0]])
+    corner_pts = np.array([[0.0, 0.0], [0.0, 2.0], [2.0, 0.0], [2.0, 2.0]])
     assert_warns(Warning, hypersphere_overlap_of_subset, corner_pts, corner_pts)