Growacq fixes on Bias construction

pycona/active_algorithms/gquacq.py

@@ -48,7 +48,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
         self.env.init_state(instance, oracle, verbose, metrics)
 
         if len(self.env.instance.bias) == 0:
-            self.env.instance.construct_bias()
+            self.env.instance.construct_bias(X)
 
         while True:
             if self.env.verbose > 0:
@@ -66,6 +66,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
                 if self.env.verbose >= 1:
                     print(f"\nLearned {self.env.metrics.cl} constraints in "
                           f"{self.env.metrics.membership_queries_count} queries.")
+                self.env.instance.bias = []
                 return self.env.instance
 
             self.env.metrics.increase_generation_time(gen_end - gen_start)

pycona/active_algorithms/growacq.py

@@ -57,7 +57,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
             Y.append(x)
             # add the constraints involving x and other added variables
             if len(self.env.instance.bias) == 0:
-                self.env.instance.construct_bias_for_var(x, Y)
+                self.env.instance.construct_bias_for_vars(x, Y)
             if verbose >= 3:
                 print(f"Added variable {x} in GrowAcq")
                 print("size of B in growacq: ", len(self.env.instance.bias))

pycona/active_algorithms/mquacq.py

@@ -39,7 +39,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
         self.env.init_state(instance, oracle, verbose, metrics)
 
         if len(self.env.instance.bias) == 0:
-            self.env.instance.construct_bias()
+            self.env.instance.construct_bias(X)
 
         while True:
             if self.env.verbose >= 3:
@@ -61,6 +61,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
                 if self.env.verbose >= 1:
                     print(f"\nLearned {self.env.metrics.cl} constraints in "
                           f"{self.env.metrics.membership_queries_count} queries.")
+                self.env.instance.bias = []
                 return self.env.instance
 
             self.env.metrics.increase_generation_time(gen_end - gen_start)

pycona/active_algorithms/mquacq2.py

@@ -53,7 +53,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
         self.cl_neighbours = np.zeros((len(self.env.instance.X), len(self.env.instance.X)), dtype=bool)
 
         if len(self.env.instance.bias) == 0:
-            self.env.instance.construct_bias()
+            self.env.instance.construct_bias(X)
 
         while True:
             gen_start = time.time()
@@ -67,6 +67,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
                 if self.env.verbose >= 1:
                     print(f"\nLearned {self.env.metrics.cl} constraints in "
                           f"{self.env.metrics.membership_queries_count} queries.")
+                self.env.instance.bias = []
                 return self.env.instance
 
             self.env.metrics.increase_generated_queries()

pycona/active_algorithms/pquacq.py

@@ -43,7 +43,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
         self.env.init_state(instance, oracle, verbose, metrics)
 
         if len(self.env.instance.bias) == 0:
-            self.env.instance.construct_bias()
+            self.env.instance.construct_bias(X)
 
         while True:
             if self.env.verbose > 0:
@@ -61,6 +61,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
                 if self.env.verbose >= 1:
                     print(f"\nLearned {self.env.metrics.cl} constraints in "
                           f"{self.env.metrics.membership_queries_count} queries.")
+                self.env.instance.bias = []
                 return self.env.instance
 
             self.env.metrics.increase_generation_time(gen_end - gen_start)

pycona/active_algorithms/quacq.py

@@ -39,7 +39,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
         self.env.init_state(instance, oracle, verbose, metrics)
 
         if len(self.env.instance.bias) == 0:
-            self.env.instance.construct_bias()
+            self.env.instance.construct_bias(X)
 
         while True:
             if self.env.verbose > 2:
@@ -57,6 +57,7 @@ def learn(self, instance: ProblemInstance, oracle: Oracle = UserOracle(), verbos
                 if self.env.verbose >= 1:
                     print(f"\nLearned {self.env.metrics.cl} constraints in "
                           f"{self.env.metrics.membership_queries_count} queries.")
+                self.env.instance.bias = []
                 return self.env.instance
 
             self.env.metrics.increase_generation_time(gen_end - gen_start)

pycona/problem_instance/problem_instance.py

@@ -217,15 +217,15 @@ def get_cpmpy_model(self):
             warnings.warn("The model is empty, as no constraint is learned yet for this instance.")
         return cp.Model(self._cl)
 
-    def construct_bias(self):
+    def construct_bias(self, X=None):
         """
         Construct the bias (candidate constraints) for the problem instance.
         """
+        if X is None:
+            X = self.X
 
         all_cons = []
 
-        X = list(self.X)
-
         for relation in self.language:
 
             abs_vars = get_scope(relation)
@@ -239,35 +239,43 @@ def construct_bias(self):
                 constraint = replace_variables(relation, replace_dict)
                 all_cons.append(constraint)
 
-        self.bias = all_cons
+        self.bias = list(set(all_cons) - set(self.cl) - set(self.excluded_cons))
+
 
-    def construct_bias_for_var(self, v1, X=None):
+    def construct_bias_for_vars(self, v1, X=None):
         """
         Construct the bias (candidate constraints) for a specific variable.
 
-        :param v1: The variable for which to construct the bias.
+        :param v1: The variable for which to construct the bias. Can also be a list of variables.
         :param X: The set of variables to consider, default is None.
         """
+
+        if not isinstance(v1, list):
+            v1 = [v1]
+
         if X is None:
             X = self.X
-        assert isinstance(X, list) and set(X).issubset(set(self.X)), "When using .construct_bias_for_var(), set parameter X must be a list of variables. Instead, got: " + str(X)
+
+        # Sort X based on variable names 
+        X = sorted(X, key=lambda var: var.name)
 
         all_cons = []
-        X = list(set(X) - {v1})
 
         for relation in self.language:
+
             abs_vars = get_scope(relation)
 
-            combs = combinations(X, len(abs_vars) - 1)
+            combs = combinations(X, len(abs_vars))
 
             for comb in combs:
-                replace_dict = {abs_vars[0]: v1}
+                replace_dict = dict()
                 for i, v in enumerate(comb):
-                    replace_dict[abs_vars[i + 1]] = v
+                    replace_dict[abs_vars[i]] = v
                 constraint = replace_variables(relation, replace_dict)
                 all_cons.append(constraint)
 
-        self.bias = all_cons
+        self.bias = [c for c in all_cons if any(v in set(get_scope(c)) for v in v1)]
+        self.bias = list(set(self.bias) - set(self.cl) - set(self.excluded_cons))
 
     def __str__(self):
         """

pycona/query_generation/pqgen.py

@@ -67,31 +67,32 @@ def blimit(self, blimit):
         """
         self._blimit = blimit
 
-    def generate(self, Y=None):
+    def reset_partial(self):
+        """
+        Reset the partial flag to False.
+        """
+        self.partial = False
+
+    def generate(self, X=None):
         """
         Generate a query using PQGen.
 
         :return: A set of variables that form the query.
         """
 
-        if Y is None:
-            Y = self.env.instance.X
-        assert isinstance(Y, list), "When generating a query, Y must be a list of variables"
-
-        # Start time (for the cutoff time)
+        if X is None:
+            X = self.env.instance.X
+        B = get_con_subset(self.env.instance.bias, X)
+        # Start time (for the cutoff t)
         t0 = time.time()
 
         # Project down to only vars in scope of B
-        Y2 = frozenset(get_variables(self.env.instance.bias))
-
-        if len(Y2) < len(Y):
-            Y = Y2
+        Y = frozenset(get_variables(B))
 
         lY = list(Y)
 
-        B = get_con_subset(self.env.instance.bias, Y)
         Cl = get_con_subset(self.env.instance.cl, Y)
-
+        
         # If no constraints left in B, just return
         if len(B) == 0:
             return set()
@@ -104,7 +105,7 @@ def generate(self, Y=None):
         if not self.partial and len(B) > self.blimit:
 
             m = cp.Model(Cl)
-            flag = m.solve()  # no time limit to ensure convergence
+            flag = m.solve(num_workers=8)  # no time limit to ensure convergence
 
             if flag and not all([c.value() for c in B]):
                 return lY
@@ -113,13 +114,16 @@ def generate(self, Y=None):
 
         m = cp.Model(Cl)
         s = cp.SolverLookup.get("ortools", m)
-
+        
         # We want at least one constraint to be violated to assure that each answer of the user
         # will lead to new information
         s += ~cp.all(B)
 
+        if self.env.verbose > 2:
+            print("Solving first without objective (to find at least one solution)...")
+
         # Solve first without objective (to find at least one solution)
-        flag = s.solve()
+        flag = s.solve(num_workers=8)
 
         t1 = time.time() - t0
         if not flag or (t1 > self.time_limit):
@@ -140,7 +144,10 @@ def generate(self, Y=None):
         # Run with the objective
         s.maximize(objective)
 
-        flag2 = s.solve(time_limit=(self.time_limit - t1))
+        if self.env.verbose > 2:
+            print("Solving with objective...")
+
+        flag2 = s.solve(time_limit=(self.time_limit - t1), num_workers=8)
 
         if flag2:
             return lY