MAINT Update submodule commit and remove cnp imports where possible (#…

…249)

* Maintenance to update submodule commit 

---------

Signed-off-by: Adam Li <[email protected]>

sktree/tree/_classes.py

@@ -6,6 +6,7 @@
 from scipy.sparse import issparse
 from sklearn.base import ClusterMixin, TransformerMixin
 from sklearn.cluster import AgglomerativeClustering
+from sklearn.utils import check_random_state
 from sklearn.utils._param_validation import Interval
 from sklearn.utils.validation import check_is_fitted
 
@@ -918,6 +919,7 @@ def _build_tree(
             Controls the randomness of the estimator.
         """
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
         _, n_features = X.shape
 
         if self.feature_combinations is None:
@@ -963,7 +965,7 @@ def _build_tree(
 
         # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
         if max_leaf_nodes < 0:
-            self.builder_ = DepthFirstTreeBuilder(
+            builder = DepthFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -972,7 +974,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
         else:
-            self.builder_ = BestFirstTreeBuilder(
+            builder = BestFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -982,7 +984,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
 
-        self.builder_.build(self.tree_, X, y, sample_weight, None)
+        builder.build(self.tree_, X, y, sample_weight, None)
 
         if self.n_outputs_ == 1:
             self.n_classes_ = self.n_classes_[0]
@@ -996,6 +998,78 @@ def _inheritable_fitted_attribute(self):
             "feature_combinations_",
         ]
 
+    def _update_tree(self, X, y, sample_weight):
+        # Update tree
+        max_leaf_nodes = -1 if self.max_leaf_nodes is None else self.max_leaf_nodes
+        min_samples_split = self.min_samples_split_
+        min_samples_leaf = self.min_samples_leaf_
+        min_weight_leaf = self.min_weight_leaf_
+        # set decision-tree model parameters
+        max_depth = np.iinfo(np.int32).max if self.max_depth is None else self.max_depth
+
+        monotonic_cst = self.monotonic_cst_
+
+        # Build tree
+        # Note: this reconstructs the builder with the same state it had during the
+        # initial fit. This is necessary because the builder is not saved as part
+        # of the class, and thus the state may be lost if pickled/unpickled.
+        criterion = self.criterion
+        if not isinstance(criterion, BaseCriterion):
+            criterion = CRITERIA_CLF[self.criterion](self.n_outputs_, self._n_classes_)
+        else:
+            # Make a deepcopy in case the criterion has mutable attributes that
+            # might be shared and modified concurrently during parallel fitting
+            criterion = copy.deepcopy(criterion)
+
+        random_state = check_random_state(self.random_state)
+
+        splitter = self.splitter
+        if issparse(X):
+            raise ValueError(
+                "Sparse input is not supported for oblique trees. "
+                "Please convert your data to a dense array."
+            )
+        else:
+            SPLITTERS = OBLIQUE_DENSE_SPLITTERS
+        if not isinstance(self.splitter, ObliqueSplitter):
+            splitter = SPLITTERS[self.splitter](
+                criterion,
+                self.max_features_,
+                min_samples_leaf,
+                min_weight_leaf,
+                random_state,
+                monotonic_cst,
+                self.feature_combinations_,
+            )
+
+        # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
+        if max_leaf_nodes < 0:
+            builder = DepthFirstTreeBuilder(
+                splitter,
+                min_samples_split,
+                min_samples_leaf,
+                min_weight_leaf,
+                max_depth,
+                self.min_impurity_decrease,
+                self.store_leaf_values,
+            )
+        else:
+            builder = BestFirstTreeBuilder(
+                splitter,
+                min_samples_split,
+                min_samples_leaf,
+                min_weight_leaf,
+                max_depth,
+                max_leaf_nodes,
+                self.min_impurity_decrease,
+                self.store_leaf_values,
+            )
+        builder.initialize_node_queue(self.tree_, X, y, sample_weight)
+        builder.build(self.tree_, X, y, sample_weight)
+
+        self._prune_tree()
+        return self
+
 
 class ObliqueDecisionTreeRegressor(SimMatrixMixin, DecisionTreeRegressor):
     """An oblique decision tree Regressor.
@@ -1785,6 +1859,7 @@ def _build_tree(
                 )
 
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
 
         # Build tree
         criterion = self.criterion
@@ -1825,7 +1900,7 @@ def _build_tree(
 
         # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
         if max_leaf_nodes < 0:
-            self.builder_ = DepthFirstTreeBuilder(
+            builder = DepthFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -1834,7 +1909,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
         else:
-            self.builder_ = BestFirstTreeBuilder(
+            builder = BestFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -1844,7 +1919,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
 
-        self.builder_.build(self.tree_, X, y, sample_weight, None)
+        builder.build(self.tree_, X, y, sample_weight, None)
 
         if self.n_outputs_ == 1:
             self.n_classes_ = self.n_classes_[0]
@@ -2263,6 +2338,7 @@ def _build_tree(
                 )
 
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
         n_samples = X.shape[0]
 
         # Build tree
@@ -2692,6 +2768,7 @@ def _build_tree(
             Controls the randomness of the estimator.
         """
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
         _, n_features = X.shape
 
         if self.feature_combinations is None:
@@ -2737,7 +2814,7 @@ def _build_tree(
 
         # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
         if max_leaf_nodes < 0:
-            self.builder_ = DepthFirstTreeBuilder(
+            builder = DepthFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -2746,7 +2823,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
         else:
-            self.builder_ = BestFirstTreeBuilder(
+            builder = BestFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -2756,7 +2833,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
 
-        self.builder_.build(self.tree_, X, y, sample_weight, None)
+        builder.build(self.tree_, X, y, sample_weight, None)
 
         if self.n_outputs_ == 1:
             self.n_classes_ = self.n_classes_[0]
@@ -3088,6 +3165,7 @@ def _build_tree(
             Controls the randomness of the estimator.
         """
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
         n_samples, n_features = X.shape
 
         if self.feature_combinations is None:

sktree/tree/_honest_tree.py

@@ -742,8 +742,10 @@ def _inherit_estimator_attributes(self):
         self.tree_ = self.estimator_.tree_
 
         # XXX: scikit-learn trees do not store their builder, or min_samples_split_
-        self.builder_ = getattr(self.estimator_, "builder_", None)
         self.min_samples_split_ = getattr(self.estimator_, "min_samples_split_", None)
+        self.min_samples_leaf_ = getattr(self.estimator_, "min_samples_leaf_", None)
+        self.min_weight_leaf_ = getattr(self.estimator_, "min_weight_leaf_", None)
+        self.monotonic_cst_ = getattr(self.estimator_, "monotonic_cst_", None)
 
     def _empty_leaf_correction(self, proba, pos=0):
         """Leaves with empty posteriors are assigned values.

sktree/tree/_marginal.pxd

@@ -3,8 +3,7 @@ import numpy as np
 cimport numpy as cnp
 
 from .._lib.sklearn.tree._tree cimport BaseTree, Node
-from .._lib.sklearn.tree._utils cimport UINT32_t
-from .._lib.sklearn.utils._typedefs cimport float32_t, float64_t, intp_t
+from .._lib.sklearn.utils._typedefs cimport float32_t, float64_t, intp_t, uint32_t
 
 
 cpdef apply_marginal_tree(

sktree/tree/_marginal.pyx

@@ -65,7 +65,7 @@ cpdef apply_marginal_tree(
     cdef intp_t n_marginals = marginal_indices.shape[0]
 
     # sklearn_rand_r random number state
-    cdef UINT32_t rand_r_state = random_state.randint(0, RAND_R_MAX)
+    cdef uint32_t rand_r_state = random_state.randint(0, RAND_R_MAX)
 
     # define a set of all marginal indices
     cdef unordered_set[intp_t] marginal_indices_map
@@ -108,7 +108,7 @@ cdef inline cnp.ndarray _apply_dense_marginal(
     unordered_set[intp_t] marginal_indices_map,
     intp_t traversal_method,
     unsigned char use_sample_weight,
-    UINT32_t* rand_r_state
+    uint32_t* rand_r_state
 ):
     """Finds the terminal region (=leaf node) for each sample in X.
 
@@ -131,7 +131,7 @@ cdef inline cnp.ndarray _apply_dense_marginal(
     use_sample_weight : unsigned char
         Whether or not to use the weighted number of samples
         in each node.
-    rand_r_state : UINT32_t
+    rand_r_state : uint32_t
         The random number state.
     """
     # Extract input

sktree/tree/_multiview.py

@@ -5,6 +5,7 @@
 
 import numpy as np
 from scipy.sparse import issparse
+from sklearn.utils import check_random_state
 from sklearn.utils._param_validation import Interval, RealNotInt, StrOptions
 
 from .._lib.sklearn.tree import DecisionTreeClassifier, _criterion
@@ -360,6 +361,7 @@ def _build_tree(
             Controls the randomness of the estimator.
         """
         monotonic_cst = None
+        self.monotonic_cst_ = monotonic_cst
         _, n_features = X.shape
 
         self.feature_combinations_ = 1
@@ -495,7 +497,7 @@ def _build_tree(
 
         # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
         if max_leaf_nodes < 0:
-            self.builder_ = DepthFirstTreeBuilder(
+            builder = DepthFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -504,7 +506,7 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
         else:
-            self.builder_ = BestFirstTreeBuilder(
+            builder = BestFirstTreeBuilder(
                 splitter,
                 min_samples_split,
                 min_samples_leaf,
@@ -514,12 +516,87 @@ def _build_tree(
                 self.min_impurity_decrease,
             )
 
-        self.builder_.build(self.tree_, X, y, sample_weight, None)
+        builder.build(self.tree_, X, y, sample_weight, None)
 
         if self.n_outputs_ == 1:
             self.n_classes_ = self.n_classes_[0]
             self.classes_ = self.classes_[0]
 
+    def _update_tree(self, X, y, sample_weight):
+        # Update tree
+        max_leaf_nodes = -1 if self.max_leaf_nodes is None else self.max_leaf_nodes
+        min_samples_split = self.min_samples_split_
+        min_samples_leaf = self.min_samples_leaf_
+        min_weight_leaf = self.min_weight_leaf_
+        # set decision-tree model parameters
+        max_depth = np.iinfo(np.int32).max if self.max_depth is None else self.max_depth
+
+        monotonic_cst = self.monotonic_cst_
+
+        # Build tree
+        # Note: this reconstructs the builder with the same state it had during the
+        # initial fit. This is necessary because the builder is not saved as part
+        # of the class, and thus the state may be lost if pickled/unpickled.
+        criterion = self.criterion
+        if not isinstance(criterion, BaseCriterion):
+            criterion = CRITERIA_CLF[self.criterion](self.n_outputs_, self._n_classes_)
+        else:
+            # Make a deepcopy in case the criterion has mutable attributes that
+            # might be shared and modified concurrently during parallel fitting
+            criterion = copy.deepcopy(criterion)
+
+        random_state = check_random_state(self.random_state)
+
+        splitter = self.splitter
+        if issparse(X):
+            raise ValueError(
+                "Sparse input is not supported for oblique trees. "
+                "Please convert your data to a dense array."
+            )
+        else:
+            SPLITTERS = DENSE_SPLITTERS
+        if not isinstance(self.splitter, ObliqueSplitter):
+            splitter = SPLITTERS[self.splitter](
+                criterion,
+                self.max_features_,
+                min_samples_leaf,
+                min_weight_leaf,
+                random_state,
+                monotonic_cst,
+                self.feature_combinations_,
+                self.feature_set_ends_,
+                self.n_feature_sets_,
+                self.max_features_per_set_,
+            )
+
+        # Use BestFirst if max_leaf_nodes given; use DepthFirst otherwise
+        if max_leaf_nodes < 0:
+            builder = DepthFirstTreeBuilder(
+                splitter,
+                min_samples_split,
+                min_samples_leaf,
+                min_weight_leaf,
+                max_depth,
+                self.min_impurity_decrease,
+                self.store_leaf_values,
+            )
+        else:
+            builder = BestFirstTreeBuilder(
+                splitter,
+                min_samples_split,
+                min_samples_leaf,
+                min_weight_leaf,
+                max_depth,
+                max_leaf_nodes,
+                self.min_impurity_decrease,
+                self.store_leaf_values,
+            )
+        builder.initialize_node_queue(self.tree_, X, y, sample_weight)
+        builder.build(self.tree_, X, y, sample_weight)
+
+        self._prune_tree()
+        return self
+
     def fit(self, X, y, sample_weight=None, check_input=True, classes=None):
         """Build a decision tree classifier from the training set (X, y).