diff --git a/therapistBNInit.py b/therapistBNInit.py
new file mode 100644
index 0000000000000000000000000000000000000000..07092251240a39b967ce0bf0b50bb83a1cd0fe6a
--- /dev/null
+++ b/therapistBNInit.py
@@ -0,0 +1,153 @@
+import random
+#attempt | user_action
+from bn_variables import User_Action, Game_State, Attempt, Agent_Assistance
+
+
+#1 create two files where to include the user model and agent model, respectively.
+def create_template(path, filename_out, filename_in):
+ dest = open(path+"/"+filename_out, "w")
+ orig = open(path+"/"+filename_in, "r")
+ contents = orig.readlines()
+ if orig.mode == 'r':
+ dest.writelines(contents)
+
+user_model_path = "/home/pal/Documents/Framework/bn_generative_model/bn_persona_model"
+user_model_filename_out = "persona_test.bif"
+user_model_filename_in = "persona_model_test.bif"
+agent_model_path = "/home/pal/Documents/Framework/bn_generative_model/bn_agent_model"
+agent_model_filename_out = "agent_test.bif"
+agent_model_filename_in = "agent_model_test.bif"
+
+create_template(path=user_model_path, filename_out=user_model_filename_out, filename_in=user_model_filename_in)
+
+user_action = [0, 1, 2]
+attempt = [0, 1, 2, 3]
+game = [0, 1, 2]
+agent_assistance = [0, 1,2 ,3 ,4 ,5 ]
+max_value_for_user = 15
+max_value_for_assistance = 30
+
+user_action_vars = ["(correct)", "(wrong)", "(timeout)"]
+agent_assistance_vars = ["(lev_0)", "(lev_1)", "(lev_2)", "(lev_3)", "(lev_4)", "(lev_5)"]
+user_action_given_agent_assistance = [[0 for i in range(User_Action.counter.value)]
+ for j in range(Agent_Assistance.counter.value)]
+attempt_given_user_action = [[0 for i in range(Attempt.counter.value)] for j in
+ range(User_Action.counter.value)]
+game_given_user_action = [[0 for i in range(Game_State.counter.value)] for j in
+ range(User_Action.counter.value)]
+
+
+attempt_preferences_for_correct_move = [2, 3, 4, 4]
+attempt_preferences_for_wrong_move = [(5-attempt_preferences_for_correct_move[a])/2 for a in range(Attempt.counter.value)]
+attempt_preferences_for_timeout = [(5-attempt_preferences_for_correct_move[a])/2 for a in range(Attempt.counter.value)]
+
+attempt_preferences_for_correct_move = [attempt_preferences_for_correct_move[i] * User_Action.counter.value for i in
+ range(len(attempt_preferences_for_correct_move))]
+attempt_preferences_for_wrong_move = [attempt_preferences_for_wrong_move[i] * User_Action.counter.value for i in
+ range(len(attempt_preferences_for_wrong_move))]
+attempt_preferences_for_timeout = [attempt_preferences_for_timeout[i] * User_Action.counter.value for i in
+ range(len(attempt_preferences_for_timeout))]
+attempt_preferences = [attempt_preferences_for_correct_move, attempt_preferences_for_wrong_move, attempt_preferences_for_timeout]
+
+user_model = open(user_model_path+"/"+user_model_filename_out, "a+")
+user_model.write("probability (game_state | user_action) { \n")
+game_preferences_for_correct_move = [2, 3, 4]
+game_preferences_for_wrong_move = [(5-game_preferences_for_correct_move[a])/2 for a in range(Game_State.counter.value)]
+game_preferences_for_timeout = [(5-game_preferences_for_correct_move[a])/2 for a in range(Game_State.counter.value)]
+
+game_preferences_for_correct_move = [game_preferences_for_correct_move[i] * User_Action.counter.value for i in
+ range(len(game_preferences_for_correct_move))]
+game_preferences_for_wrong_move = [game_preferences_for_wrong_move[i] * User_Action.counter.value for i in
+ range(len(game_preferences_for_wrong_move))]
+game_preferences_for_timeout = [game_preferences_for_timeout[i] * User_Action.counter.value for i in
+ range(len(game_preferences_for_timeout))]
+game_preferences = [game_preferences_for_correct_move, game_preferences_for_wrong_move, game_preferences_for_timeout]
+
+for elem in range(len(game_given_user_action)):
+ game_preferences[elem] = list(map(lambda x:x/sum(game_preferences[elem]), game_preferences[elem]))
+ user_model.write(str(user_action_vars[elem]) + "\t" +
+ str(game_preferences[elem][0]) + "," +
+ str(game_preferences[elem][1]) + "," +
+ str(game_preferences[elem][2]) + "; \n")
+user_model.write("}\n")
+
+
+user_model.write("probability (attempt | user_action) { \n")
+for elem in range(len(attempt_given_user_action)):
+ attempt_preferences[elem] = list(map(lambda x:x/sum(attempt_preferences[elem]), attempt_preferences[elem]))
+ user_model.write(str(user_action_vars[elem])+ "\t" +
+ str(attempt_preferences[elem][0])+","+
+ str(attempt_preferences[elem][1])+","+
+ str(attempt_preferences[elem][2]) + "," +
+ str(attempt_preferences[elem][3])+"; \n")
+user_model.write("}\n")
+
+
+
+user_model.write("probability (user_action | agent_assistance) { \n")
+assistance_score_for_correct_move = [2, 3, 3, 4, 5, 5]
+assistance_score_for_wrong_move = [(5 - assistance_score_for_correct_move[i]) / 2 for i in
+ range(len(assistance_score_for_correct_move))]
+assistance_score_for_timeout = [(5 - assistance_score_for_correct_move[i]) / 2 for i in
+ range(len(assistance_score_for_correct_move))]
+assistance_score_for_correct_move = [assistance_score_for_correct_move[i] * User_Action.counter.value for i in
+ range(len(assistance_score_for_correct_move))]
+assistance_score_for_wrong_move = [assistance_score_for_wrong_move[i] * User_Action.counter.value for i in
+ range(len(assistance_score_for_wrong_move))]
+assistance_score_for_timeout = [assistance_score_for_timeout[i] * User_Action.counter.value for i in
+ range(len(assistance_score_for_timeout))]
+
+for elem in range(Agent_Assistance.counter.value):
+ #setting up the effect of each level on the user_action
+ den = (assistance_score_for_correct_move[elem]+assistance_score_for_wrong_move[elem]+assistance_score_for_timeout[elem])
+ user_action_given_agent_assistance[elem] = [assistance_score_for_correct_move[elem]/den, assistance_score_for_wrong_move[elem]/den,
+ assistance_score_for_timeout[elem]/den]
+ user_model.write(str(agent_assistance_vars[elem])+ "\t"+
+ str(user_action_given_agent_assistance[elem][0])+","+
+ str(user_action_given_agent_assistance[elem][1])+","+
+ str(user_action_given_agent_assistance[elem][2])+"; \n")
+user_model.write("}")
+
+
+create_template(path=agent_model_path, filename_out=agent_model_filename_out, filename_in=agent_model_filename_in )
+
+agent_assistance_given_game_attempt = [[0 for ass in range(Agent_Assistance.counter.value)]
+ for g in range(Game_State.counter.value)
+ for a in range(Attempt.counter.value)]
+agent_model = open(agent_model_path+"/"+agent_model_filename_out, "a+")
+
+attempt_game_vars = [
+"(beg, att_1)", "(beg, att_2)","(beg, att_3)","(beg, att_4)",
+"(mid, att_1)", "(mid, att_2)","(mid, att_3)","(mid, att_4)",
+"(end, att_1)", "(end, att_2)","(end, att_3)","(end, att_4)",
+]
+
+agent_model.write("probability (agent_assistance | game_state, attempt) { \n")
+assistance_preferences = [[4, 5, 2, 1, 1, 1],
+ [1, 3, 5, 3, 1, 1],
+ [1, 2, 2, 5, 4, 1],
+ [1, 1, 2, 5, 4, 4]]
+
+it = 0
+for g in range(Game_State.counter.value):
+ for a in range(Attempt.counter.value):
+ print("it:", it)
+ #get the preference
+ assistance_preferences = [[(assistance_preferences[i][j] * Agent_Assistance.counter.value) for j in range(len(assistance_preferences[i]))] for i in range(len(assistance_preferences))]
+ agent_assistance_given_game_attempt[it] = list(map(lambda x:x/sum(assistance_preferences[a]), assistance_preferences[a]))
+ agent_model.write(str(attempt_game_vars[it])+"\t"+
+ str(agent_assistance_given_game_attempt[it][0])+", "+
+ str(agent_assistance_given_game_attempt[it][1]) + ", " +
+ str(agent_assistance_given_game_attempt[it][2]) + ", " +
+ str(agent_assistance_given_game_attempt[it][3]) + ", " +
+ str(agent_assistance_given_game_attempt[it][4]) + ", " +
+ str(agent_assistance_given_game_attempt[it][5]) + "; \n"
+ )
+ it += 1
+agent_model.write("}")
+
+
+print(attempt_given_user_action)
+print(game_given_user_action)
+print(user_action_given_agent_assistance)
+print(agent_assistance_given_game_attempt)
\ No newline at end of file