Struct Gpu 

pub struct Gpu;

Sealed marker for the group-backed payload of a TypedTensor.

TypedTensor<T, R, Gpu> carries provider allocation authority, identity and retirement state through its allocation group, so host access stays fallible and must be prepared or mapped explicitly. The marker is not proof of a particular device ordinal or provider.

Examples

use tenferro_tensor::{DynRank, Gpu, Host, TypedTensor};

let host = TypedTensor::<f64, DynRank, Host>::from_host_vec_col_major(vec![2], vec![1.0, 2.0])?;
let gpu: TypedTensor<f64, DynRank, Gpu> = host.promote()?;
assert_eq!(gpu.host_data()?, &[1.0, 2.0]);

Trait Implementations

impl Debug for Gpu

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

Formats the value using the given formatter.

impl Representation for Gpu