Variables
int	K = 10

int	N = 2000

int	M = 1000

list	bandits = []

	single_bandit = bandit.Normal(k=K)

list	agents

list	agent_names

	rewards = numpy.zeros(shape=(len(agents), N, M), dtype=numpy.float)

	_table

	a

float	cumulative_mean_reward = 0.0

	action = test_agent.act()

list	reward = bandits[n].select(index=action)

	mean_rewards = numpy.mean(a=rewards, axis=1)

Variable Documentation

◆ _table

analysis._table

protected

Definition at line 45 of file analysis.py.

◆ a

analysis.a

Definition at line 45 of file analysis.py.

◆ action

analysis.action = test_agent.act()

Definition at line 49 of file analysis.py.

◆ agent_names

list analysis.agent_names

Initial value:

=  [
    '0.0',
    '0.01',
    '0.1',
]

Definition at line 33 of file analysis.py.

◆ agents

list analysis.agents

Initial value:

=  [
    agent.Greedy(k=K),
    agent.EpsilonGreedy(k=K, epsilon=0.01),
    agent.EpsilonGreedy(k=K, epsilon=0.1),
]

Definition at line 28 of file analysis.py.

◆ bandits

list analysis.bandits = []

Definition at line 24 of file analysis.py.

◆ cumulative_mean_reward

float analysis.cumulative_mean_reward = 0.0

Definition at line 46 of file analysis.py.

◆ K

int analysis.K = 10

Definition at line 17 of file analysis.py.

◆ M

int analysis.M = 1000

Definition at line 21 of file analysis.py.

◆ mean_rewards

analysis.mean_rewards = numpy.mean(a=rewards, axis=1)

Definition at line 58 of file analysis.py.

◆ N

int analysis.N = 2000

Definition at line 19 of file analysis.py.

◆ reward

analysis.reward = bandits[n].select(index=action)

Definition at line 50 of file analysis.py.

◆ rewards

analysis.rewards = numpy.zeros(shape=(len(agents), N, M), dtype=numpy.float)

Definition at line 39 of file analysis.py.

◆ single_bandit

analysis.single_bandit = bandit.Normal(k=K)

Definition at line 26 of file analysis.py.

Variables