Add optimized shuffle_do() method to AgentSet

benchmarks/BoltzmannWealth/boltzmann_wealth.py

@@ -60,7 +60,7 @@ def __init__(self, seed=None, n=100, width=10, height=10):
 
     def step(self):
         """Run the model for a single step."""
-        self.agents.shuffle().do("step")
+        self.agents.shuffle_do("step")
         # collect data
         self.datacollector.collect(self)
 

benchmarks/WolfSheep/wolf_sheep.py

@@ -227,8 +227,8 @@ def __init__(
 
     def step(self):
         """Run one step of the model."""
-        self.agents_by_type[Sheep].shuffle(inplace=True).do("step")
-        self.agents_by_type[Wolf].shuffle(inplace=True).do("step")
+        self.agents_by_type[Sheep].shuffle_do("step")
+        self.agents_by_type[Wolf].shuffle_do("step")
 
 
 if __name__ == "__main__":

mesa/agent.py

@@ -293,6 +293,19 @@ def do(self, method: str | Callable, *args, **kwargs) -> AgentSet:
 
         return self
 
+    def shuffle_do(self, method: str | Callable, *args, **kwargs) -> AgentSet:
+        agents = list(self._agents.keys())
+        self.random.shuffle(agents)
+
+        if isinstance(method, str):
+            for agent in agents:
+                getattr(agent, method)(*args, **kwargs)
+        else:
+            for agent in agents:
+                method(agent, *args, **kwargs)
+
+        return self
+
     def map(self, method: str | Callable, *args, **kwargs) -> list[Any]:
         """Invoke a method or function on each agent in the AgentSet and return the results.
 

mesa/experimental/devs/examples/epstein_civil_violence.py

@@ -293,7 +293,7 @@ def __init__(
 
     def step(self):
         """Run one step of the model."""
-        self.active_agents.shuffle(inplace=True).do("step")
+        self.active_agents.shuffle_do("step")
 
 
 if __name__ == "__main__":

mesa/experimental/devs/examples/wolf_sheep.py

@@ -230,8 +230,8 @@ def __init__(
 
     def step(self):
         """Perform one step of the model."""
-        self.agents_by_type[Sheep].shuffle(inplace=True).do("step")
-        self.agents_by_type[Wolf].shuffle(inplace=True).do("step")
+        self.agents_by_type[Sheep].shuffle_do("step")
+        self.agents_by_type[Wolf].shuffle_do("step")
 
 
 if __name__ == "__main__":

tests/test_agent.py

@@ -437,6 +437,42 @@ def test_agentset_map_callable():
     assert all(i == entry for i, entry in zip(results, range(1, 11)))
 
 
+def test_agentset_shuffle_do():
+    """Test AgentSet.shuffle_do method."""
+    model = Model()
+
+    class TestAgentShuffleDo(Agent):
+        def __init__(self, model):
+            super().__init__(model)
+            self.called = False
+
+        def test_method(self):
+            self.called = True
+
+    agents = [TestAgentShuffleDo(model) for _ in range(100)]
+    agentset = AgentSet(agents, model)
+
+    # Test shuffle_do with a string method name
+    agentset.shuffle_do("test_method")
+    assert all(agent.called for agent in agents)
+
+    # Reset the called flag
+    for agent in agents:
+        agent.called = False
+
+    # Test shuffle_do with a callable
+    agentset.shuffle_do(lambda agent: setattr(agent, "called", True))
+    assert all(agent.called for agent in agents)
+
+    # Verify that the order is indeed shuffled
+    original_order = list(agentset)
+    shuffled_order = []
+    agentset.shuffle_do(lambda agent: shuffled_order.append(agent))
+    assert (
+        original_order != shuffled_order
+    ), "The order should be different after shuffle_do"
+
+
 def test_agentset_get_attribute():
     """Test AgentSet.get for attributes."""
     model = Model()