Module training 

Training utilities and infrastructure

This module provides comprehensive utilities for training neural networks, including advanced features like gradient accumulation, mixed precision training, distributed training, and sophisticated training loop management.

Re-exports

pub use gradient_accumulation::*;

pub use gradient_checkpointing::*;

pub use mixed_precision::*;

pub use progress_monitor::*;

pub use quantization_aware::*;

pub use sparse_training::*;

Modules

gradient_accumulation

Gradient accumulation utilities

gradient_checkpointing

Gradient checkpointing utilities

mixed_precision

Mixed precision training utilities

progress_monitor

Progress monitoring utilities

quantization_aware

Quantization-aware training utilities

sparse_training

Sparse training utilities

Structs

TrainingConfig

Configuration structure for training neural networks

TrainingSession

Training session for tracking training history

ValidationSettings

Configuration for validation during training