def test_passreward(env_name='PerceptualDecisionMaking-v0',
num_steps=1000,
verbose=False):
"""
Test pass-reward wrapper.
TODO: explain wrapper
Parameters
----------
env_name : str, optional
enviroment to wrap.. The default is 'PerceptualDecisionMaking-v0'.
num_steps : int, optional
number of steps to run the environment (1000)
verbose : boolean, optional
whether to print observation and reward (False)
Returns
-------
None.
"""
env = gym.make(env_name)
env = PassReward(env)
obs = env.reset()
for stp in range(num_steps):
action = env.action_space.sample()
obs, rew, done, info = env.step(action)
assert obs[-1] == rew, 'Previous reward is not part of observation'
if verbose:
print(obs)
print(rew)
print('--------')
if done:
env.reset()
def test_passreward(env_name, num_steps=10000, verbose=False, **envArgs):
env = gym.make(env_name, **envArgs)
env = PassReward(env)
obs = env.reset()
for stp in range(num_steps):
action = env.action_space.sample()
obs, rew, done, info = env.step(action)
if verbose:
print(obs)
print(rew)
print('--------')
if done:
env.reset()