MNT Refactor tree splitter to use memoryviews (#23273)

glemaitre · glemaitre · commit ad5936f8d2d4 · 2022-08-05T14:58:55.000+02:00
Co-authored-by: Guillaume Lemaitre &lt;g.lemaitre58@gmail.com&gt;
diff --git a/doc/whats_new/v1.2.rst b/doc/whats_new/v1.2.rst
@@ -65,6 +65,13 @@ Changelog
   now conducts validation for `max_features` and `feature_names_in` parameters.
   :pr:`23299` by :user:`Long Bao <lorentzbao>`.
 
+:mod:`sklearn.tree`
+...................
+
+- |Fix| Fixed invalid memory access bug during fit in
+  :class:`tree.DecisionTreeRegressor` and :class:`tree.DecisionTreeClassifier`.
+  :pr:`23273` by `Thomas Fan`_.
+
 Code and Documentation Contributors
 -----------------------------------
 
diff --git a/sklearn/tree/_splitter.pxd b/sklearn/tree/_splitter.pxd
@@ -46,13 +46,13 @@ cdef class Splitter:
     cdef object random_state             # Random state
     cdef UINT32_t rand_r_state           # sklearn_rand_r random number state
 
-    cdef SIZE_t* samples                 # Sample indices in X, y
+    cdef SIZE_t[::1] samples             # Sample indices in X, y
     cdef SIZE_t n_samples                # X.shape[0]
     cdef double weighted_n_samples       # Weighted number of samples
-    cdef SIZE_t* features                # Feature indices in X
-    cdef SIZE_t* constant_features       # Constant features indices
+    cdef SIZE_t[::1] features            # Feature indices in X
+    cdef SIZE_t[::1] constant_features   # Constant features indices
     cdef SIZE_t n_features               # X.shape[1]
-    cdef DTYPE_t* feature_values         # temp. array holding feature values
+    cdef DTYPE_t[::1] feature_values     # temp. array holding feature values
 
     cdef SIZE_t start                    # Start position for the current node
     cdef SIZE_t end                      # End position for the current node
diff --git a/sklearn/tree/_splitter.pyx b/sklearn/tree/_splitter.pyx
