def test_invalid_not_run_plot(self):
rng = np.random.RandomState(seed=7)
sim = Simulator(bandits=[("example",
MAB([0, 1],
LearningPolicy.EpsilonGreedy()))],
decisions=[rng.randint(0, 2) for _ in range(10)],
rewards=[rng.randint(0, 100) for _ in range(10)],
contexts=[[rng.rand() for _ in range(5)]
for _ in range(10)],
scaler=StandardScaler(),
test_size=0.4,
batch_size=0,
is_ordered=True,
seed=7)
with self.assertRaises(AssertionError):
sim.plot()
runtime = (end - start) / 60
print('Complete', str(runtime) + ' minutes')
print('\n')
for mab_name, mab in sim.bandits:
print(mab_name)
# Since simulation is offline, print the bandit stats directly
print('Worst Case Scenario', sim.bandit_to_arm_to_stats_min[mab_name])
print('Average Case Scenario', sim.bandit_to_arm_to_stats_avg[mab_name])
print('Best Case Scenario:', sim.bandit_to_arm_to_stats_max[mab_name])
print('\n\n')
sim.plot('max', True)
####################################
# Context-Free Simulation
####################################
start = time()
sim = Simulator(context_free_mabs, decisions, rewards, contexts=None,
scaler=None, test_size=0.5, is_ordered=False, batch_size=100, seed=123456)
sim.run()
end = time()
runtime = (end - start) / 60
print('Complete', str(runtime) + ' minutes')
print('\n')
for mab_name, mab in sim.bandits: