Add node cutoff

CHANGELOG.md

@@ -2,6 +2,7 @@
 
 ## Unreleased
 ### Added
+- Added cutoffNode and test
 ### Fixed
 ### Changed
 ### Removed

src/pyscipopt/scip.pxd

@@ -877,6 +877,7 @@ cdef extern from "scip/scip.h":
     SCIP_Real SCIPgetPrimalbound(SCIP* scip)
     SCIP_Real SCIPgetGap(SCIP* scip)
     int SCIPgetDepth(SCIP* scip)
+    SCIP_RETCODE SCIPcutoffNode(SCIP* scip, SCIP_NODE* node)
     SCIP_Bool SCIPhasPrimalRay(SCIP * scip)
     SCIP_Real SCIPgetPrimalRayVal(SCIP * scip, SCIP_VAR * var)
     SCIP_RETCODE SCIPaddSolFree(SCIP* scip, SCIP_SOL** sol, SCIP_Bool* stored)

src/pyscipopt/scip.pxi

@@ -272,7 +272,7 @@
     if rc == SCIP_OKAY:
         pass
     elif rc == SCIP_ERROR:
         raise Exception('SCIP: unspecified error!')
     elif rc == SCIP_NOMEMORY:
         raise MemoryError('SCIP: insufficient memory error!')
     elif rc == SCIP_READERROR:
@@ -1313,7 +1313,7 @@
 
         """
         return SCIPnodeIsActive(self.scip_node)
-
+    
     def isPropagatedAgain(self):
         """
         Is the node marked to be propagated again?
@@ -2335,6 +2335,16 @@
         """
         return SCIPgetDepth(self._scip)
 
+    def cutoffNode(self, Node node):
+        """
+        marks node and whole subtree to be cut off from the branch and bound tree.
+
+        Parameters
+        ----------
+        node : Node
+        """
+        PY_SCIP_CALL( SCIPcutoffNode(self._scip, node.scip_node) )
+
     def infinity(self):
         """
         Retrieve SCIP's infinity value.

tests/helpers/utils.py

@@ -1,7 +1,7 @@
 from pyscipopt import Model, quicksum, SCIP_PARAMSETTING, exp, log, sqrt, sin
 from typing import List
 
-def random_mip_1(disable_sepa=True, disable_huer=True, disable_presolve=True, node_lim=2000, small=False):
+def random_mip_1(disable_sepa=True, disable_heur=True, disable_presolve=True, node_lim=2000, small=False):
     model = Model()
 
     x0 = model.addVar(lb=-2, ub=4)
@@ -41,7 +41,7 @@ def random_mip_1(disable_sepa=True, disable_huer=True, disable_presolve=True, no
 
     if disable_sepa:
         model.setSeparating(SCIP_PARAMSETTING.OFF)
-    if disable_huer:
+    if disable_heur:
         model.setHeuristics(SCIP_PARAMSETTING.OFF)
     if disable_presolve:
         model.setPresolve(SCIP_PARAMSETTING.OFF)

tests/test_heur.py

@@ -124,7 +124,7 @@ def inner():
 
 def test_simple_round_heur():
     # create solver instance
-    s = random_mip_1(disable_sepa=False, disable_huer=False, node_lim=1)
+    s = random_mip_1(disable_sepa=False, disable_heur=False, node_lim=1)
     heuristic = SimpleRoundingHeuristic()
     s.includeHeur(heuristic, "SimpleRounding", "simple rounding heuristic implemented in python", "Y",
                   timingmask=SCIP_HEURTIMING.DURINGLPLOOP)

tests/test_node.py

@@ -0,0 +1,21 @@
+from pyscipopt import SCIP_RESULT, Eventhdlr, SCIP_EVENTTYPE
+from helpers.utils import random_mip_1
+
+class cutoffEventHdlr(Eventhdlr):
+    def eventinit(self):
+        self.model.catchEvent(SCIP_EVENTTYPE.NODEFOCUSED, self)
+
+    def eventexec(self, event):
+        self.model.cutoffNode(self.model.getCurrentNode())
+        return {'result': SCIP_RESULT.SUCCESS}
+
+def test_cutoffNode():
+    m = random_mip_1(disable_heur=True, disable_presolve=True, disable_sepa=True)
+
+    hdlr = cutoffEventHdlr()
+
+    m.includeEventhdlr(hdlr, "test", "test")
+
+    m.optimize()
+
+    assert m.getNSols() == 0

tests/test_nogil.py

@@ -6,7 +6,7 @@
 
 
 def test_optimalNogil():
-    ori_model = random_mip_1(disable_sepa=False, disable_huer=False, disable_presolve=False, node_lim=2000, small=True) 
+    ori_model = random_mip_1(disable_sepa=False, disable_heur=False, disable_presolve=False, node_lim=2000, small=True) 
     models = [Model(sourceModel=ori_model) for _ in range(N_Threads)]
     for i in range(N_Threads):
         models[i].setParam("randomization/permutationseed", i)

tests/test_solution.py

@@ -197,7 +197,7 @@ def test_getSols():
 
 
 def test_getOrigin_retrasform():
-    m = random_mip_1(disable_sepa=False, disable_huer=False, disable_presolve=False, small=True)
+    m = random_mip_1(disable_sepa=False, disable_heur=False, disable_presolve=False, small=True)
     m.optimize()
 
     sol = m.getBestSol()
@@ -208,11 +208,11 @@ def test_getOrigin_retrasform():
 
 
 def test_translate():
-    m = random_mip_1(disable_sepa=False, disable_huer=False, disable_presolve=False, small=True)
+    m = random_mip_1(disable_sepa=False, disable_heur=False, disable_presolve=False, small=True)
     m.optimize()
     sol = m.getBestSol()
 
-    m1 = random_mip_1(disable_sepa=False, disable_huer=False, disable_presolve=False, small=True)
+    m1 = random_mip_1(disable_sepa=False, disable_heur=False, disable_presolve=False, small=True)
     sol1 = sol.translate(m1)
     assert m1.addSol(sol1) == True
     assert m1.getNSols() == 1

tests/test_strong_branching.py

@@ -140,7 +140,7 @@ def branchexeclp(self, allowaddcons):
 
 
 def test_strong_branching():
-    scip = random_mip_1(disable_presolve=False, disable_huer=False, small=True, node_lim=500)
+    scip = random_mip_1(disable_presolve=False, disable_heur=False, small=True, node_lim=500)
 
     strong_branch_rule = StrongBranchingRule(scip, idempotent=False)
     scip.includeBranchrule(strong_branch_rule, "strong branch rule", "custom strong branching rule",
@@ -155,7 +155,7 @@ def test_strong_branching():
 
 
 def test_strong_branching_idempotent():
-    scip = random_mip_1(disable_presolve=False, disable_huer=False, small=True, node_lim=500)
+    scip = random_mip_1(disable_presolve=False, disable_heur=False, small=True, node_lim=500)
 
     strong_branch_rule = StrongBranchingRule(scip, idempotent=True)
     scip.includeBranchrule(strong_branch_rule, "strong branch rule", "custom strong branching rule",
@@ -170,7 +170,7 @@ def test_strong_branching_idempotent():
 
 
 def test_dummy_feature_selector():
-    scip = random_mip_1(disable_presolve=False, disable_huer=False, small=True, node_lim=300)
+    scip = random_mip_1(disable_presolve=False, disable_heur=False, small=True, node_lim=300)
 
     feature_dummy_branch_rule = FeatureSelectorBranchingRule(scip)
     scip.includeBranchrule(feature_dummy_branch_rule, "dummy branch rule", "custom feature creation branching rule",