Struct relearn::torch::history::features::PackedHistoryFeatures[][src]

pub struct PackedHistoryFeatures {
    pub discount_factor: f64,
    pub episode_ranges: Vec<Range<usize>>,
    pub batch_sizes: Option<Vec<i64>>,
    pub batch_sizes_tensor: Option<Tensor>,
    pub observation_features: Option<Tensor>,
    pub actions: Option<Tensor>,
    pub returns: Option<Tensor>,
    pub rewards: Option<Tensor>,
    pub device: Device,
}
Expand description

Packed history features.

Panics

The PackedHistoryFeaturesView this provides will panic if the requested features is not available.

Fields

discount_factor: f64episode_ranges: Vec<Range<usize>>batch_sizes: Option<Vec<i64>>batch_sizes_tensor: Option<Tensor>observation_features: Option<Tensor>actions: Option<Tensor>returns: Option<Tensor>rewards: Option<Tensor>device: Device

Trait Implementations

Formats the value using the given formatter. Read more

Packed rewards. A 1D f32 tensor.

Discount factor for calculating returns.

Episode index ranges sorted in decreasing order of episode length.

Batch sizes in the packing. Read more

Batch sizes in the packing. A 1D i64 tensor.

Packed observation features. A 2D f64 tensor.

Packed action values. Read more

Packed returns (discounted reward-to-go). A 1D f32 tensor.

Device on which tensors will be placed.

This method tests for self and other values to be equal, and is used by ==. Read more

This method tests for !=.

Auto Trait Implementations

Blanket Implementations

Gets the TypeId of self. Read more

Immutably borrows from an owned value. Read more

Mutably borrows from an owned value. Read more

Performs the conversion.

Performs the conversion.

The type returned in the event of a conversion error.

Performs the conversion.

The type returned in the event of a conversion error.

Performs the conversion.