Library providing sublinear-time solvers for large-scale math problems like PageRank, graph Laplacians, and AI attention using 8 algorithms including Neumann Series, Conjugate Gradient, Forward/Backward Push, and more. Optimized for scale with SIMD SpMV, fused kernels, and arena allocators; supports WASM and NAPI bindings.
Зареждане на още......
На тази страница
ruvector-solver
Solves problems faster as data grows, powering real-time self-learning.
| Algorithm | Complexity | Best For |
|---|---|---|
| Neumann Series | O(k · nnz) | Diagonally dominant |
| Conjugate Gradient | O(√κ · log(1/ε) · nnz) | SPD systems |
| Forward Push | O(1/ε) | Single-source PageRank |
| Backward Push | O(1/ε) | Reverse relevance |
| Hybrid Random Walk | O(√n/ε) | Pairwise relevance |
| TRUE | O(log n) amortized | Laplacian systems |
| BMSSP | O(nnz · log n) | Multigrid solve |
| Auto Router | Automatic | Optimal selection |
Key optimizations: AVX2 SIMD, fused residuals, bounds-check elimination.
cargo add ruvector-solver --features all-algorithms
Free and open-source.
Разгледайте още елементи, свързани с този