PerArmPolicyInfo(log_probability, predicted_rewards_mean, predicted_rewards_optimistic, predicted_rewards_sampled, bandit_policy_type, chosen_arm_features)
tf_agents.bandits.policies.policy_utilities.PerArmPolicyInfo(
log_probability=(), predicted_rewards_mean=(), predicted_rewards_optimistic=(),
predicted_rewards_sampled=(), bandit_policy_type=(), chosen_arm_features=()
)
Attributes |
log_probability
|
|
predicted_rewards_mean
|
|
predicted_rewards_optimistic
|
|
predicted_rewards_sampled
|
|
bandit_policy_type
|
|
chosen_arm_features
|
|