added clarifying comments

treeple/stats/forest.py

@@ -171,13 +171,12 @@ def build_coleman_forest(
         )
 
         # if we are returning the posteriors, then we need to replace the
-        # sparse indices and values with an array
+        # sparse indices and values with an array. We convert the sparse data
+        # to dense data, so that the function returns results in a consistent format.
         if return_posteriors:
             n_trees = y_pred_proba_orig_perm.shape[0] // 2
             n_samples = y_pred_proba_orig_perm.shape[1]
-            # slicing a csc matrix this way is not efficient, but it is
-            # it is only done once, so I am not sure if it is worth it to
-            # optimize this.
+
             to_coords_data = lambda x: (x.row.astype(int), x.col.astype(int), x.data)
 
             row, col, data = to_coords_data(y_pred_proba_orig_perm[:n_trees, :].tocoo())

treeple/stats/tests/test_forest.py

@@ -250,6 +250,8 @@ def test_build_coleman_forest(use_bottleneck: bool, use_sparse: bool):
     elif use_bottleneck and utils.DISABLE_BN_ENV_VAR in os.environ:
         del os.environ[utils.DISABLE_BN_ENV_VAR]
 
+    # We need to reload the modules after changing the environment variable
+    # because an environment variable is used to disable bottleneck
     importlib.reload(utils)
     importlib.reload(stats)