scirs2_series/
neural_forecasting.rs

1//! Neural Forecasting Models for Time Series
2//!
3//! This module provides cutting-edge implementations for neural network-based
4//! time series forecasting, including LSTM, GRU, Transformer, Mamba/State Space Models,
5//! Temporal Fusion Transformers, and Mixture of Experts architectures.
6//! These implementations focus on core algorithmic components and can be
7//! extended with actual neural network frameworks.
8//!
9//! ## Advanced Architectures
10//! - **LSTM Networks**: Long Short-Term Memory networks for sequence modeling
11//! - **Transformer Models**: Self-attention based architectures
12//! - **N-BEATS**: Neural basis expansion analysis for time series forecasting
13//! - **Mamba/State Space Models**: Linear complexity for long sequences with selective state spaces
14//! - **Flash Attention**: Memory-efficient attention computation for transformers
15//! - **Temporal Fusion Transformers**: Specialized architecture for time series forecasting
16//! - **Mixture of Experts**: Conditional computation for model scaling
17
18// Re-export all module components for backward compatibility and ease of use
19pub use self::{
20    attention::*, config::*, lstm::*, mamba::*, mixture_of_experts::*, nbeats::*,
21    temporal_fusion::*, transformer::*,
22};
23
24// Module declarations
25pub mod attention;
26pub mod config;
27pub mod lstm;
28pub mod mamba;
29pub mod mixture_of_experts;
30pub mod nbeats;
31pub mod temporal_fusion;
32pub mod transformer;
scirs2_series/neural_forecasting.rs

scirs2_series/
neural_forecasting.rs
