added submodel function

docs/getting_started/simulator.ipynb

@@ -2,17 +2,18 @@
   "cells": [
     {
       "cell_type": "code",
-      "execution_count": 14,
+      "execution_count": null,
       "id": "a8ddc37c-66d2-43e4-8162-6be19a1d70a1",
       "metadata": {},
       "outputs": [],
       "source": [
-        "from stormvogel import visualization, show, simulator, model"
+        "from stormvogel import visualization, show, simulator\n",
+        "import stormvogel.model"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 15,
+      "execution_count": null,
       "id": "cab40f99-3460-4497-8b9f-3d669eee1e11",
       "metadata": {},
       "outputs": [],
@@ -85,27 +86,10 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 16,
+      "execution_count": null,
       "id": "eb0fadc0-7bb6-4c1d-ae3e-9e16527726ab",
       "metadata": {},
-      "outputs": [
-        {
-          "name": "stdout",
-          "output_type": "stream",
-          "text": [
-            "ModelType.MDP with name None\n",
-            "\n",
-            "States:\n",
-            "State 0 with labels ['init'] and features {}\n",
-            "State 1 with labels ['carchosen'] and features {}\n",
-            "State 2 with labels ['open'] and features {}\n",
-            "State 3 with labels ['goatrevealed'] and features {}\n",
-            "State 4 with labels ['target', 'done'] and features {}\n",
-            "\n",
-            "Transitions:\n"
-          ]
-        }
-      ],
+      "outputs": [],
       "source": [
         "#we want to simulate this model. That is, we start at the initial state and then\n",
         "#we walk through the model according to transition probabilities.\n",
@@ -125,29 +109,12 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 17,
+      "execution_count": null,
       "id": "59ac1e34-866c-42c4-b19b-c2a15c830e2e",
       "metadata": {
         "scrolled": true
       },
-      "outputs": [
-        {
-          "name": "stdout",
-          "output_type": "stream",
-          "text": [
-            "ModelType.MDP with name None\n",
-            "\n",
-            "States:\n",
-            "State 0 with labels ['init'] and features {}\n",
-            "State 1 with labels ['carchosen'] and features {}\n",
-            "State 2 with labels ['open'] and features {}\n",
-            "State 3 with labels ['goatrevealed'] and features {}\n",
-            "State 4 with labels ['done'] and features {}\n",
-            "\n",
-            "Transitions:\n"
-          ]
-        }
-      ],
+      "outputs": [],
       "source": [
         "#it still chooses random actions but we can prevent this by providing a scheduler:\n",
         "taken_actions = {}\n",
@@ -161,74 +128,35 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 20,
+      "execution_count": null,
       "id": "22871288-755c-463f-9150-f207c2f5c211",
       "metadata": {},
-      "outputs": [
-        {
-          "data": {
-            "application/vnd.jupyter.widget-view+json": {
-              "model_id": "f92e1a66e2034c70832b36e2bd7fe70f",
-              "version_major": 2,
-              "version_minor": 0
-            },
-            "text/plain": [
-              "Output()"
-            ]
-          },
-          "metadata": {},
-          "output_type": "display_data"
-        },
-        {
-          "data": {
-            "application/vnd.jupyter.widget-view+json": {
-              "model_id": "b413c4c87cb34895b0a4551b4e49ed16",
-              "version_major": 2,
-              "version_minor": 0
-            },
-            "text/plain": [
-              "HBox(children=(Output(), Output()))"
-            ]
-          },
-          "metadata": {},
-          "output_type": "display_data"
-        },
-        {
-          "data": {
-            "text/plain": [
-              "<stormvogel.visualization.Visualization at 0x7f6e782fe4b0>"
-            ]
-          },
-          "execution_count": 20,
-          "metadata": {},
-          "output_type": "execute_result"
-        }
-      ],
+      "outputs": [],
       "source": [
         "#we can also visualize the partial model that we get from the simulator:\n",
-        "show.show(partial_model)"
+        "show.show(partial_model)\n"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 19,
+      "execution_count": null,
       "id": "34d0c293-d090-4e3d-9e80-4351f5fcba62",
       "metadata": {},
-      "outputs": [
-        {
-          "name": "stdout",
-          "output_type": "stream",
-          "text": [
-            "initial state --(action: empty)--> state: 2 --(action: open0)--> state: 7 --(action: empty)--> state: 17 --(action: stay)--> state: 33\n"
-          ]
-        }
-      ],
+      "outputs": [],
       "source": [
         "#we can also use another simulator function that returns a path instead of a partial model:\n",
         "path = stormvogel.simulator.simulate_path(mdp, steps=4, scheduler=scheduler, seed=123456)\n",
         "\n",
         "print(path)"
       ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "99c763fa-82ea-42ff-8833-79c640f14518",
+      "metadata": {},
+      "outputs": [],
+      "source": []
     }
   ],
   "metadata": {

stormvogel/model.py

@@ -4,6 +4,7 @@
 from enum import Enum
 from fractions import Fraction
 from typing import cast
+import copy
 
 Parameter = str
 
@@ -391,6 +392,13 @@ def supports_observations(self):
         """Returns whether this model supports observations."""
         return self.type == ModelType.POMDP
 
+    def get_sub_model(self, states: list[State]) -> "Model":
+        sub_model = copy.deepcopy(self)
+        for state in states:
+            sub_model.delete_state(state, normalize=True, reassign_ids=True)
+
+        return sub_model
+
     def is_well_defined(self) -> bool:
         """Checks if all sums of outgoing transition probabilities for all states equal 1"""
 

stormvogel/simulator.py

@@ -277,4 +277,7 @@ def get_range_index(stateid: int):
                 if simulator.is_done():
                     break
 
-    return partial_model
+    # TODO: refactor
+    states = list(partial_model.states.values())
+    sub_model = model.get_sub_model(states)
+    return sub_model

tests/test_simulator.py

@@ -34,6 +34,8 @@ def test_simulate():
     for stateid in other_dtmc.states.keys():
         rewardmodel3.rewards[stateid] = float(1)
 
+    print(partial_model)
+
     assert partial_model == other_dtmc
 
     # we make a monty hall mdp and run the simulator with it