burn-tensor 0.20.1

Tensor library with user-friendly APIs and automatic differentiation support
Documentation
#![cfg_attr(not(feature = "std"), no_std)]
#![warn(missing_docs)]
#![cfg_attr(docsrs, feature(doc_cfg))]

//! This library provides the core abstractions required to run tensor operations with Burn.
//! `Tensor`s are generic over the backend to allow users to perform operations using different `Backend` implementations.
//! Burn's tensors also support auto-differentiation thanks to the `AutodiffBackend` trait.

#[macro_use]
extern crate derive_new;

extern crate alloc;

mod tensor;

pub(crate) use tensor::check::macros::check;
pub use tensor::*;

// Re-exported types
pub use burn_backend::{AllocationProperty, Bytes, StreamId, bf16, f16, read_sync, try_read_sync};