//! # RustyTorch++ Classical ML Library
//!
//! GPU-accelerated classical machine learning algorithms with sklearn-compatible APIs.
//!
//! ## Features
//! - GPU-accelerated implementations using CUDA kernels
//! - sklearn-compatible APIs for easy migration
//! - Zero-copy tensor operations via rtx-tensor
//! - Parallel training with rayon
//!
//! ## Modules
//! - `trees`: Decision trees, random forests, gradient boosting
//! - `linear`: Linear models with various regularization
//! - `clustering`: K-means, DBSCAN clustering algorithms
//! - `bayesian`: Naive Bayes, Gaussian Process models
//! - `neighbors`: K-nearest neighbors implementations
//!
//! ## Example
//! ```rust
//! # use rtx_ml_classic::trees::DecisionTree;
//! # use rtx_tensor::{Tensor, Device};
//! let device = Device::cuda(0).unwrap();
//! let x = Tensor::randn([100, 4], &device);
//! let y = Tensor::randint(0, 2, [100], &device);
//!
//! let mut tree = DecisionTree::new()
//!     .max_depth(5)
//!     .criterion("gini");
//! tree.fit(&x, &y).unwrap();
//! let predictions = tree.predict(&x).unwrap();
//! ```

pub mod bayesian;
pub mod clustering;
pub mod error;
pub mod linear;
pub mod neighbors;
pub mod trees;

// Re-export main error types
pub use error::{MLError, Result};

#[cfg(test)]
mod tests {
    use super::*;

    /// Smoke test: ensure basic module structure works
    #[test]
    fn test_crate_compiles() {
        assert!(true);
    }
}