Crate tenferro_cpu 

CPU backend, kernels, provider selection, and CPU resource pools.

Examples

use tenferro_cpu::{add, CpuBackend};
use tenferro_tensor::{Tensor, TensorBackend, TensorElementwise};

let mut backend = CpuBackend::new();
let a = Tensor::from_vec_col_major(vec![2], vec![1.0_f64, 2.0])?;
let b = Tensor::from_vec_col_major(vec![2], vec![3.0_f64, 4.0])?;
let c = backend.add(&a, &b)?;
assert_eq!(c.as_slice::<f64>().unwrap(), &[4.0, 6.0]);
let direct = add(&a, &b)?;
assert_eq!(direct.shape(), &[2]);

Re-exports

pub use affinity::available_parallelism;

pub use affinity::process_cpu_affinity_count;

pub use backend::CpuBackend;

pub use backend::CpuBackendKind;

pub use context::CpuContext;

Modules

affinity

backend

context

Structs

BufferPoolStats

Snapshot of typed host buffers retained by a [BufferPool].

Functions

abs

Compute elementwise absolute values.

add

Add two CPU tensors elementwise.

broadcast_in_dim

clamp

Clamp CPU tensor values elementwise between lower and upper bounds.

compare

Compare two CPU tensors elementwise.

conj

Conjugate a real or complex CPU tensor elementwise.

convert

Convert a tensor to another dtype using checked dtype conversion.

div

Divide two CPU tensors elementwise.

dynamic_slice

dynamic_update_slice

Return operand with update written at dynamic starts.

embed_diagonal

extract_diagonal

gather

maximum

Compute elementwise maximum values.

minimum

Compute elementwise minimum values.

mul

Multiply two CPU tensors elementwise.

neg

Negate a CPU tensor elementwise.

pad

reduce_max

reduce_min

reduce_prod

reduce_sum

reshape

scatter

select

Select values from two tensors using a boolean predicate tensor.

sign

Compute elementwise signs.

transpose

tril

triu