steffen/marl-factory-grid
1
0
Fork 0
mirror of https://github.com/illiumst/marl-factory-grid.git synced 2025-12-14 03:00:37 +01:00
Code Issues Packages Projects Releases Wiki Activity

Reword TSP runner

Browse Source
This commit is contained in:
Julian Schönberger
2024-08-09 16:36:39 +02:00
parent 0a8eaf8827
commit c56bb3f188
1 changed files with 37 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

41
marl_factory_grid/algorithms/tsp/TSP_runner.py
View File

@@ -1,4 +1,5 @@
import os import os
import pickle
from pathlib import Path from pathlib import Path
from tqdm import trange from tqdm import trange
@@ -8,14 +9,14 @@ from marl_factory_grid.algorithms.tsp.contortions import get_dirt_quadrant_tsp_a
def dirt_quadrant_multi_agent_tsp_eval(emergent_phenomenon): def dirt_quadrant_multi_agent_tsp_eval(emergent_phenomenon):
run_tsp_setting("dirt_quadrant", emergent_phenomenon) run_tsp_setting("dirt_quadrant", emergent_phenomenon, log=False)
def two_rooms_multi_agent_tsp_eval(emergent_phenomenon): def two_rooms_multi_agent_tsp_eval(emergent_phenomenon):
run_tsp_setting("two_rooms", emergent_phenomenon) run_tsp_setting("two_rooms", emergent_phenomenon, log=False)
def run_tsp_setting(config_name, emergent_phenomenon, n_episodes=1): def run_tsp_setting(config_name, emergent_phenomenon, n_episodes=1, log=False):
# Render at each step? # Render at each step?
render = True render = True
@@ -35,8 +36,13 @@ def run_tsp_setting(config_name, emergent_phenomenon, n_episodes=1):
with open(f"{results_path}/env_config.txt", "w") as txt_file: with open(f"{results_path}/env_config.txt", "w") as txt_file:
txt_file.write(str(factory.conf)) txt_file.write(str(factory.conf))
still_existing_dirt_piles = []
reached_flags = []
for episode in trange(n_episodes): for episode in trange(n_episodes):
_ = factory.reset() _ = factory.reset()
still_existing_dirt_piles.append([])
reached_flags.append([])
done = False done = False
if render: if render:
factory.render() factory.render()
@@ -48,6 +54,7 @@ def run_tsp_setting(config_name, emergent_phenomenon, n_episodes=1):
else: else:
print("Config name does not exist. Abort...") print("Config name does not exist. Abort...")
break break
ep_steps = 0
while not done: while not done:
a = [x.predict() for x in agents] a = [x.predict() for x in agents]
# Have this condition, to terminate as soon as all dirt piles are collected. This ensures that the implementation # Have this condition, to terminate as soon as all dirt piles are collected. This ensures that the implementation
@@ -55,7 +62,33 @@ def run_tsp_setting(config_name, emergent_phenomenon, n_episodes=1):
if 'DirtPiles' in list(factory.state.entities.keys()) and factory.state.entities['DirtPiles'].global_amount == 0.0: if 'DirtPiles' in list(factory.state.entities.keys()) and factory.state.entities['DirtPiles'].global_amount == 0.0:
break break
obs_type, _, _, done, info = factory.step(a) obs_type, _, _, done, info = factory.step(a)
if 'DirtPiles' in list(factory.state.entities.keys()):
still_existing_dirt_piles[-1].append(len(factory.state.entities['DirtPiles']))
if 'Destinations' in list(factory.state.entities.keys()):
reached_flags[-1].append(sum([1 for ele in [x.was_reached() for x in factory.state['Destinations']] if ele]))
ep_steps += 1
if render: if render:
factory.render() factory.render()
if done: if done:
break break
cleaned_dirt_piles_per_step = []
if 'DirtPiles' in list(factory.state.entities.keys()):
for ep in still_existing_dirt_piles:
cleaned_dirt_piles_per_step.append([max(ep)-ep[idx] for idx, value in enumerate(ep)])
# Remove first element and add last element where all dirt piles have been collected
del cleaned_dirt_piles_per_step[-1][0]
cleaned_dirt_piles_per_step[-1].append(max(still_existing_dirt_piles[-1]))
# Add last entry to reached_flags
print(ep_steps)
print(reached_flags)
print(cleaned_dirt_piles_per_step)
if log:
if 'DirtPiles' in list(factory.state.entities.keys()):
metrics_data = {"cleaned_dirt_piles_per_step": cleaned_dirt_piles_per_step}
if 'Destinations' in list(factory.state.entities.keys()):
metrics_data = {"reached_flags": reached_flags}
with open(f"{results_path}/metrics", "wb") as pickle_file:
pickle.dump(metrics_data, pickle_file)