Update submodule

Signed-off-by: Adam Li <adam2392@gmail.com>
neurodata · Aug 16, 2023 · 565e4d6 · 565e4d6
1 parent d5ce01c
commit 565e4d6
Show file tree

Hide file tree

Showing 3 changed files with 2 additions and 57 deletions.
diff --git a/sktree/tree/_classes.py b/sktree/tree/_classes.py
@@ -1630,7 +1630,7 @@ def _build_tree(
                 )
 
         if self.data_dims is None:
-            self.data_dims_ = np.array((1, X.shape[1]), dtype=np.intp)
+            self.data_dims_ = np.array((1, X.shape[1]), dtype=np.int8)
         else:
             if np.prod(self.data_dims) != X.shape[1]:
                 raise RuntimeError(f"Data dimensions {self.data_dims} do not match {X.shape[1]}.")

diff --git a/sktree/tree/manifold/_morf_splitter.pyx b/sktree/tree/manifold/_morf_splitter.pyx
@@ -36,7 +36,6 @@ cdef class PatchSplitter(BaseObliqueSplitter):
         const DOUBLE_t[:, ::1] y,
         const DOUBLE_t[:] sample_weight,
         const unsigned char[::1] missing_values_in_feature_mask,
-        # const INT32_t[:] n_categories
     ) except -1:
         BaseObliqueSplitter.init(self, X, y, sample_weight, missing_values_in_feature_mask)
 
@@ -198,60 +197,6 @@ cdef class BestPatchSplitter(BaseDensePatchSplitter):
                 self.feature_weight.base if self.feature_weight is not None else None,
             ), self.__getstate__())
 
-    # def __getstate__(self):
-    #     """Getstate re-implementation, for pickling."""
-    #     d = {}
-    #     # capacity is inferred during the __setstate__ using nodes
-    #     d["criterion"] = self.criterion
-    #     d["max_features"] = self.max_features
-    #     d["min_samples_leaf"] = self.min_samples_leaf
-    #     d["min_weight_leaf"] = self.min_weight_leaf
-    #     d['random_state'] = self.random_state
-    #     d['monotonic_cst'] = self.monotonic_cst.base
-    #     d['min_patch_dims'] = self.min_patch_dims.base
-    #     d['max_patch_dims'] = self.max_patch_dims.base
-    #     d['dim_contiguous'] = self.dim_contiguous.base
-    #     d['data_dims'] = self.data_dims.base
-    #     d['boundary'] = self.boundary
-    #     d['feature_weight'] = self.feature_weight.base
-    #     return d
-
-    # def __setstate__(self, d):
-    #     self.criterion = d["criterion"]
-    #     self.max_features = d["max_features"]
-    #     self.min_samples_leaf = d["min_samples_leaf"]
-    #     self.min_weight_leaf = d["min_weight_leaf"]
-    #     self.random_state = d['random_state']
-    #     self.monotonic_cst = d['monotonic_cst']
-
-    #     self.min_patch_dims = d['min_patch_dims']
-    #     self.max_patch_dims = d['max_patch_dims']
-    #     self.dim_contiguous = d['dim_contiguous']
-    #     self.data_dims = d['data_dims']
-    #     self.boundary = d['boundary']
-    #     self.feature_weight = d['feature_weight']
-
-    #     whether or not to perform some discontinuous sampling
-    #     if not all(self.dim_contiguous):
-    #         self._discontiguous = True
-    #     else:
-    #         self._discontiguous = False
-
-    #     # Sparse max_features x n_features projection matrix
-    #     self.proj_mat_weights = vector[vector[DTYPE_t]](self.max_features)
-    #     self.proj_mat_indices = vector[vector[SIZE_t]](self.max_features)
-
-    #     # initialize state to allow generalization to higher-dimensional tensors
-    #     self.ndim = self.data_dims.shape[0]
-
-    #     # create a buffer for storing the patch dimensions sampled per projection matrix
-    #     self.patch_dims_buff = np.zeros(self.data_dims.shape[0], dtype=np.intp)
-    #     self.unraveled_patch_point = np.zeros(self.data_dims.shape[0], dtype=np.intp)
-
-    #     # # initialize a buffer to allow for Fisher-Yates
-    #     self._index_patch_buffer = np.zeros(np.max(self.max_patch_dims), dtype=np.intp)
-    #     self._index_data_buffer = np.zeros(np.max(self.data_dims), dtype=np.intp)
-
     cdef (SIZE_t, SIZE_t) sample_top_left_seed(self) noexcept nogil:
         """Sample the top-left seed for the n-dim patch.
 

diff --git a/sktree/tree/tests/test_tree.py b/sktree/tree/tests/test_tree.py
@@ -179,7 +179,7 @@ def test_pickle_splitters():
     [
         ObliqueDecisionTreeClassifier(random_state=12),
         ObliqueDecisionTreeRegressor(random_state=12),
-        PatchObliqueDecisionTreeClassifier(random_state=12),
+        # PatchObliqueDecisionTreeClassifier(random_state=12),
         PatchObliqueDecisionTreeRegressor(random_state=12),
     ]
 )