Crate axonml_core 

Axonml Core - Foundation Layer for the Axonml ML Framework

This crate provides the core abstractions that underpin the entire Axonml machine learning framework. It handles device management, memory storage, data types, and backend implementations.

Key Features

• Device abstraction (CPU, CUDA, Vulkan, Metal, WebGPU)
• Type-safe data type system (f32, f64, f16, i32, i64, bool)
• Efficient memory storage with reference counting
• Pluggable backend architecture

Example

use axonml_core::{Device, DType, Storage};

let device = Device::Cpu;
let storage = Storage::<f32>::zeros(1024, device);

@version 0.1.0 @author AutomataNexus Development Team

Re-exports

pub use allocator::Allocator;

pub use allocator::DefaultAllocator;

pub use device::Device;

pub use dtype::DType;

pub use dtype::Float;

pub use dtype::Numeric;

pub use dtype::Scalar;

pub use error::Error;

pub use error::Result;

pub use storage::Storage;

Modules

allocator

Allocator - Memory Allocation Traits and Implementations

backends

Backends - Device-Specific Implementations

device

Device Abstraction - Hardware Backend Management

dtype

Data Types - Axonml Type System

error

Error Types - Axonml Core Error Handling

prelude

Convenient imports for common usage.

storage

Storage - Raw Memory Management for Tensors