def buildEnvironment(explored_states, state_dict, G, G2, G_disrupted,
ActionList, supply_nodes):
Schedule = []
first_state_id = random.choice(explored_states)
remaining_demand = sum(state_dict[(first_state_id, 'demand')]) > 0
first_state = st.State(state_dict[(first_state_id, 'debris')],
state_dict[(first_state_id, 'supply')],
state_dict[(first_state_id, 'demand')],
state_dict[(first_state_id, 'resource')],
first_state_id)
# This is not the ordered schedule but rather the roads cleared until this state
cleared_roads = [
i for i, val in enumerate(state_dict[(first_state_id, 'debris')])
if val == 0
]
Schedule.extend(cleared_roads)
for cr in cleared_roads:
ed = [edge for edge, edge_id in ActionList.items() if edge_id == cr][0]
G2[ed[0]][ed[1]]['debris'] = 0
G_disrupted.add_edge(ed[0], ed[1])
reachable_nodes = [] # list for reachable nodes
for s in supply_nodes:
reachable_nodes.extend(list(nx.dfs_preorder_nodes(G_disrupted, s)))
actions = funcs2.initializeActionSpace(
reachable_nodes, G, ActionList
) # actions are the admissable action indices corresponding in ActionList
return first_state, actions, Schedule, reachable_nodes
betw_centrality_regular = {}
betw_centrality_debris = {}
betw_centrality_regular_sp = {}
explored_states = []
state_dict[(0, 'demand')] = initial_demand
state_dict[(0, 'debris')] = initial_debris
state_dict[(0, 'supply')] = initial_supply
state_dict[(0, 'period')] = 1
state_dict[(0, 'resource')] = 0
initial_state = st.State(initial_debris, initial_supply, initial_demand, 0, None)
id_counter, id_dict = initial_state.getStateIndex(id_counter, id_dict)
actions = funcs2.initializeActionSpace(supply_nodes, G, ActionList)
explored_states.append(initial_state.ID)
for e in range(int(n_episodes)):
# Initialize the environment
#remaining_demand = True
Schedule = [] #Initialization - these roads are as if they are cleared before
reachable_nodes = set(supply_nodes)
rem_resource = resource
period = 1
G_restored = nx.Graph()
G_restored.add_nodes_from(range(n_nodes))
phi_sa = {}
betw_centrality_service = {}
betw_centrality_regular = {}
betw_centrality_debris = {}
betw_centrality_regular_sp = {}
state_dict[(0, 'demand')] = initial_demand
state_dict[(0, 'debris')] = initial_debris
state_dict[(0, 'supply')] = initial_supply
state_dict[(0, 'period')] = 1
state_dict[(0, 'resource')] = 0
initial_state = st.State(initial_debris, initial_supply, initial_demand, 0,
None)
id_counter, id_dict = initial_state.getStateIndex(id_counter, id_dict)
actions = funcs2.initializeActionSpace(supply_nodes, G, ActionList)
explored_states.append(initial_state.ID)
step_size = 0.1
# Get the actual optimal calculated to see if GD is working
df_q = pd.read_csv(
'C:/Users/ulusan.a/Desktop/RL_rep/RL/data_files/Q_optimalVI_INS8.csv',
sep=',')
df_q.set_index('Unnamed: 0', inplace=True)
df = pd.read_csv(
'C:/Users/ulusan.a/Desktop/RL_rep/RL/data_files/basis_INS8.csv', sep=',')
df.set_index('Unnamed: 0', inplace=True)
# q_column = pd.DataFrame(index=df.index.copy(), columns=['q_optimal', 'q_pred'])