Struct PolicyGradientConfig 

pub struct PolicyGradientConfig<T: Float + Debug + Send + Sync + 'static> {
    pub base_config: RLOptimizerConfig<T>,
    pub method: PolicyGradientMethod,
    pub ppo_config: PPOConfig<T>,
    pub trpo_config: TRPOConfig<T>,
    pub policy_scheduler: Option<RLScheduler<T>>,
    pub value_scheduler: Option<RLScheduler<T>>,
    pub use_baseline: bool,
    pub importance_sampling: bool,
    pub max_is_ratio: T,
}

Policy gradient optimizer configuration

Fields

base_config: RLOptimizerConfig<T>

Base RL configuration

method: PolicyGradientMethod

Policy gradient method

ppo_config: PPOConfig<T>

PPO-specific parameters

trpo_config: TRPOConfig<T>

TRPO-specific parameters

policy_scheduler: Option<RLScheduler<T>>

Learning rate scheduler for policy

value_scheduler: Option<RLScheduler<T>>

Learning rate scheduler for value function

use_baseline: bool

Use baseline (value function) for variance reduction

importance_sampling: bool

Enable importance sampling for off-policy updates

max_is_ratio: T

Maximum importance sampling ratio

Trait Implementations

impl<T: Clone + Float + Debug + Send + Sync + 'static> Clone for PolicyGradientConfig<T>

fn clone(&self) -> PolicyGradientConfig<T>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<T: Debug + Float + Debug + Send + Sync + 'static> Debug for PolicyGradientConfig<T>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<T: Float + Debug + Send + Sync + 'static + FromPrimitive> Default for PolicyGradientConfig<T>

fn default() -> Self

Returns the “default value” for a type.