def test_unparsed_args():
with pytest.raises(ValueError, match='Unparsed arguments'):
run(
shlex.split(
'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --agent rlai.agents.q_value.EpsilonGreedy --epsilon 0.2 0.0 --testing'
))
def test_plot():
# without pdf (without random seed)
run(
shlex.split(
'--T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --agent rlai.agents.q_value.EpsilonGreedy --epsilon 0.2 0.0 --plot --figure-name test'
))
# with pdf
run(
shlex.split(
f'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --agent rlai.agents.q_value.EpsilonGreedy --epsilon 0.2 0.0 --plot --pdf-save-path {tempfile.NamedTemporaryFile(delete=False).name}'
))
def test_mancala():
monitors = run(
shlex.split(
f'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.mancala.Mancala --initial-count 4 --agent {dump_agent()}'
))
# uncomment the following line and run test to update fixture
# with open(f'{os.path.dirname(__file__)}/fixtures/test_mancala.pickle', 'wb') as file:
# pickle.dump(monitors, file)
with open(f'{os.path.dirname(__file__)}/fixtures/test_mancala.pickle',
'rb') as file:
monitors_fixture = pickle.load(file)
assert_monitors(monitors, monitors_fixture)
def test_k_armed_bandit_epsilon_greedy_optimistic():
monitors = run(
shlex.split(
'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --agent rlai.agents.q_value.EpsilonGreedy --epsilon 0.0 --initial-q-value 5 --alpha 0.1'
))
# uncomment the following line and run test to update fixture
# with open(f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_epsilon_greedy_optimistic.pickle', 'wb') as file:
# pickle.dump(monitors, file)
with open(
f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_epsilon_greedy_optimistic.pickle',
'rb') as file:
monitors_fixture = pickle.load(file)
assert_monitors(monitors, monitors_fixture)
def test_gamblers_problem():
monitors = run(
shlex.split(
f'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.gamblers_problem.GamblersProblem --p-h 0.4 --agent {dump_agent()} --log INFO'
))
# uncomment the following line and run test to update fixture
# with open(f'{os.path.dirname(__file__)}/fixtures/test_gamblers_problem.pickle', 'wb') as file:
# pickle.dump(monitors, file)
with open(
f'{os.path.dirname(__file__)}/fixtures/test_gamblers_problem.pickle',
'rb') as file:
monitors_fixture = pickle.load(file)
assert_monitors(monitors, monitors_fixture)
def test_k_armed_bandit_preference_gradient_without_baseline():
monitors = run(
shlex.split(
'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --q-star-mean 4 --agent rlai.agents.h_value.PreferenceGradient --step-size-alpha 0.1'
))
# uncomment the following line and run test to update fixture
# with open(f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_preference_gradient_without_baseline.pickle', 'wb') as file:
# pickle.dump(monitors, file)
with open(
f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_preference_gradient_without_baseline.pickle',
'rb') as file:
monitors_fixture = pickle.load(file)
assert_monitors(monitors, monitors_fixture)
def test_k_armed_bandit_upper_confidence_bound():
monitors = run(
shlex.split(
'--random-seed 12345 --T 100 --n-runs 200 --environment rlai.environments.bandit.KArmedBandit --k 10 --agent rlai.agents.q_value.UpperConfidenceBound --c 0 1'
))
# uncomment the following line and run test to update fixture
# with open(f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_upper_confidence_bound.pickle', 'wb') as file:
# pickle.dump(monitors, file)
with open(
f'{os.path.dirname(__file__)}/fixtures/test_k_armed_bandit_upper_confidence_bound.pickle',
'rb') as file:
monitors_fixture = pickle.load(file)
assert_monitors(monitors, monitors_fixture)
