forked from AIRI-Institute/AriGraph
-
Notifications
You must be signed in to change notification settings - Fork 0
/
pipeline_arigraph.py
232 lines (183 loc) · 9.74 KB
/
pipeline_arigraph.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
import json
from time import time
from agents.parent_agent import GPTagent
from graphs.contriever_graph import ContrieverGraph
from utils.envs_cfg import ENV_NAMES, FIRST_OBS, MAIN_GOALS
from utils.win_cond import win_cond_clean_place, win_cond_clean_take
from utils.textworld_adapter import TextWorldWrapper, graph_from_facts
from prompts.system_prompts import default_system_prompt, system_plan_agent, \
system_action_agent_sub_expl, if_exp_prompt
from utils.utils import Logger, observation_processing, find_unexplored_exits, \
simulate_environment_actions, action_processing, action_deprocessing
# Changeable part of pipeline
log_file = "arigraph_nav4_without_episodic"
# env_name can be picked from:
# ["hunt", "hunt_hard", "cook", "cook_hard", "cook_hardest", "cook_rl_baseline", "clean"]
# for test another envs edit utils.envs_cfg
env_name = "hunt_hard"
model = "gpt-4o"
retriever_device = "cpu"
api_key = "insert your key here"
n_prev, topk_episodic = 5, 2
max_steps, n_attempts = 150, 1
need_exp = True
# End of changeable part of pipeline
main_goal = MAIN_GOALS[env_name]
log = Logger(log_file)
env = TextWorldWrapper(ENV_NAMES[env_name])
agent = GPTagent(model = model, system_prompt=default_system_prompt, api_key = api_key)
agent_plan = GPTagent(model = "gpt-4-0125-preview", system_prompt=system_plan_agent, api_key = api_key)
agent_action = GPTagent(model = "gpt-4-0125-preview", system_prompt=system_action_agent_sub_expl, api_key = api_key)
agent_if_expl = GPTagent(model = model, system_prompt=if_exp_prompt, api_key = api_key)
def run():
total_amount, total_time = 0, 0
for attempt in range(n_attempts):
log("\n\n\n\n\n\n\nAttempt: " + str(attempt + 1))
log("=" * 70)
observations, history = [], []
locations = set()
observation, info = env.reset()
action = "start"
plan0 = f'''{{
"main_goal": {main_goal},
"plan_steps": [
{{
"sub_goal_1": "Start the game",
"reason": "You should start the game"
}},
],
}}'''
subgraph = []
previous_location = observation_processing(env.curr_location).lower()
attempt_amount, attempt_time = 0, 0
done = False
graph = ContrieverGraph(model, system_prompt = "You are a helpful assistant", device = retriever_device, api_key = api_key)
reward, step_reward = 0, 0
rewards = []
for step in range(max_steps):
start = time()
log("Step: " + str(step + 1))
observation = observation.split("$$$")[-1]
observation = observation_processing(observation)
if step == 0:
observation += FIRST_OBS[env_name]
observation = "Game step #" + str(step + 1) + "\n" + observation
inventory = env.get_inventory()
if done:
log("Game itog: " + observation)
log("\n" * 10)
break
log("Observation: " + observation)
log("Inventory: " + str(inventory))
locations.add(observation_processing(env.curr_location).lower())
observed_items, _ = agent.item_processing_scores(observation, plan0)
items = {key.lower(): value for key, value in observed_items.items()}
log("Crucial items: " + str(items))
subgraph, top_episodic = graph.update(observation, observations, plan=plan0, prev_subgraph=subgraph, locations=list(locations), curr_location=observation_processing(env.curr_location).lower(), previous_location=previous_location, action=action, log=log, items1 = items, topk_episodic=topk_episodic)
observation += f"\nInventory: {inventory}"
log("Length of subgraph: " + str(len(subgraph)))
log("Associated triplets: " + str(subgraph))
log("Episodic memory: " + str(top_episodic))
if_explore, _ = agent_if_expl.generate(prompt=f"Plan: \n{plan0}", t=0.2) if need_exp else ("False", 0)
if_explore = "True" in if_explore
log('If explore: ' + str(if_explore))
#Exploration
all_unexpl_exits = get_unexpl_exits(locations, graph) if if_explore else ""
if if_explore:
log(all_unexpl_exits)
valid_actions = [action_processing(action) for action in env.get_valid_actions()] + env.expand_action_space() if "cook" in env_name else env.get_valid_actions()
valid_actions += [f"go to {loc}" for loc in locations]
log("Valid actions: " + str(valid_actions))
hist_obs = "\n".join(history)
plan0 = planning(hist_obs, observation, plan0, subgraph, top_episodic, if_explore, all_unexpl_exits)
action = choose_action(hist_obs, observation, subgraph, top_episodic, plan0, all_unexpl_exits, valid_actions, if_explore)
observations.append(observation)
observations = observations[-n_prev:]
history.append(f"Observation: {observation}\nAction taken: {action}")
history = history[-n_prev:]
previous_location = observation_processing(env.curr_location).lower()
observation, step_reward, done, info = process_action_get_reward(action, env, info, graph, locations, env_name)
reward += step_reward
rewards.append(reward)
step_amount = agent.total_amount + graph.total_amount + agent_plan.total_amount + agent_action.total_amount + agent_if_expl.total_amount - total_amount
attempt_amount += step_amount
total_amount += step_amount
log(f"\nTotal amount: {round(total_amount, 2)}$, attempt amount: {round(attempt_amount, 2)}$, step amount: {round(step_amount, 2)}$")
step_time = time() - start
attempt_time += step_time
total_time += step_time
log(f"Total time: {round(total_time, 2)} sec, attempt time: {round(attempt_time, 2)} sec, step time: {round(step_time, 2)} sec")
log("=" * 70)
log(f"\n\nTOTAL REWARDS: {rewards}\n\n")
def process_action_get_reward(action, env, info, graph, locations, env_name):
G_true = graph_from_facts(info)
full_graph = G_true.edges(data = True)
step_reward = 0
is_nav = "go to" in action
done = False
if is_nav:
destination = action.split('go to ')[1]
path = graph.find_path(observation_processing(env.curr_location).lower(), destination, locations)
if not isinstance(path, list):
observation = path
else:
log("\n\nNAVIGATION\n\n")
for hidden_step, hidden_action in enumerate(path):
observation, reward_, done, info = env.step(hidden_action)
step_reward += reward_
if done:
break
log("Navigation step: " + str(hidden_step + 1))
log("Observation: " + observation + "\n\n")
else:
observation, reward_, done, info = env.step(action)
step_reward += reward_
G_true_new = graph_from_facts(info)
full_graph_new = G_true_new.edges(data = True)
step_reward = simulate_environment_actions(full_graph, full_graph_new, win_cond_clean_take, win_cond_clean_place) \
if env_name == "clean" else step_reward
return observation, step_reward, done, info
def choose_action(observations, observation, subgraph, top_episodic, plan0, all_unexpl_exits, valid_actions, if_explore):
# \n5. Your {topk_episodic} most relevant episodic memories from the past for the current situation: {top_episodic}.
prompt = f'''\n1. Main goal: {main_goal}
\n2. History of {n_prev} last observations and actions: {observations}
\n3. Your current observation: {observation}
\n4. Information from the memory module that can be relevant to current situation: {subgraph}
\n5. Your {topk_episodic} most relevant episodic memories from the past for the current situation: {top_episodic}.
\n6. Your current plan: {plan0}'''
if if_explore:
prompt += f'''\n7. Yet unexplored exits in the environment: {all_unexpl_exits}'''
prompt += f'''\n\nPossible actions in current situation: {valid_actions}'''
t = 0.2 if need_exp else 1
action0, cost_action = agent_action.generate(prompt, jsn=True, t = t)
log("Action: " + action0)
try:
action_json = json.loads(action0)
action = action_json["action_to_take"]
except:
log("!!!INCORRECT ACTION CHOICE!!!")
action = "look"
action = action_deprocessing(action) if "cook" in env_name else action
return action
def planning(observations, observation, plan0, subgraph, top_episodic, if_explore, all_unexpl_exits):
# \n5. Your {topk_episodic} most relevant episodic memories from the past for the current situation: {top_episodic}.
prompt = f'''\n1. Main goal: {main_goal}
\n2. History of {n_prev} last observations and actions: {observations}
\n3. Your current observation: {observation}
\n4. Information from the memory module that can be relevant to current situation: {subgraph}
\n5. Your {topk_episodic} most relevant episodic memories from the past for the current situation: {top_episodic}.
\n6. Your previous plan: {plan0}'''
if if_explore:
prompt += f'''\n7. Yet unexplored exits in the environment: {all_unexpl_exits}'''
plan0, cost_plan = agent_plan.generate(prompt, jsn=True, t=0.2)
log("Plan0: " + plan0)
return plan0
def get_unexpl_exits(locations, graph):
all_unexpl_exits = ""
for loc in locations:
loc_gr = graph.get_associated_triplets([loc], steps = 1)
unexplored_exits = find_unexplored_exits(loc, loc_gr)
all_unexpl_exits += f'\nUnexplored exits for {loc}: {unexplored_exits}'
return all_unexpl_exits
if __name__ == "__main__":
run()