gavindiaz/optimiz-rs
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
90 Commits 3 Branches 4 Tags
0463382fcb3d8561b232ce1963bdbfe7e20beb93
Commit Graph
4 Commits
Author SHA1 Message Date
ThotDjehuty 24556f51d7 release(v2.0.0): finalize PyPI metadata, README rewrite, v2 benchmark + McKean-Vlasov animation 
- Restore correct PyPI distribution name 'optimiz-rs' (continuity with v1.0.x).
  Rust crate stays 'optimiz-rs'; Python module is 'optimizr'.
- python/optimizr/__init__.py:
  * Bump __version__ from stale '0.2.0' to '2.0.0'.
  * Eagerly bind every v2 primitive from _core (so dir(optimizr), IDE
    auto-complete and 'from optimizr import X' all work without relying on
    the lazy __getattr__ fallback).
  * Extend __all__ with 38 new v2 entries.
- README.md: full v2 features section grouped by domain (rough volatility,
  BSDE/PDE, stochastic control, mean-field, topology/graphs/signatures,
  risk/robust inference, point processes, Kalman). Embedded
  examples/mckean_vlasov.gif at the top. Added v2 benchmark table.
- examples/benchmark_v2.py: honest benchmark vs pure-Python/NumPy
  references on intrinsically loopy workloads. Best-of-3, single-thread,
  Apple M2: HMM 67.7x, DE 13.9x, signatures 11.2x, Hawkes 3.3x, MCMC 1.7x.
- examples/animate_mckean_vlasov.py + examples/mckean_vlasov.gif (5MB):
  cinematic 800-particle mean-reverting McKean-Vlasov flow animation
  using optimizr.mean_reverting_mckean_vlasov.
- tests/test_v2_api.py already in place: 20/20 pass.
 2026-05-14 21:54:49 +02:00
ThotDjehuty 1799a2fa7b release(v2.0.0): promote alpha to stable, fix PyPI naming, add v2 non-regression suite 
- Bump Cargo.toml + pyproject.toml from 2.0.0-alpha.1 to 2.0.0.
- Restore PyPI distribution name to 'optimizr' (continuity with v1.4.x).
  Rust crate stays 'optimiz-rs'; both expose Python module 'optimizr'.
- README: new 'What's New in v2.0.0' section listing every advertised
  primitive (solve_volterra, solve_fractional_ode, linear_bsde_constant_coeffs,
  mean_reverting_mckean_vlasov, historical_var_py, etc.) with corrected
  install command 'pip install optimizr'.
- tests/test_v2_api.py: 20-test non-regression suite with analytic
  ground-truth checks for every v2 primitive plus a parametrised guard
  over the v1.x public surface.
- CHANGELOG: 2.0.0 entry documenting the release.

Build verification:
- maturin develop --release --features python-bindings -> optimizr-2.0.0 wheel built
- pytest tests/test_v2_api.py: 20 passed, 0 failed
- cargo test --lib --no-default-features: 124 passed; 5 pre-existing failures
  (mrsjd, ou_estimator, hurst_random_walk, hjb_solver_symmetry, regime_switching)
  unchanged since v1.1.
 2026-05-14 14:00:39 +02:00
ThotDjehuty d6b6018b9c release(v2.0.0-alpha.1): top-level reorg + bsde/pde/stochastic_control + mean_field/agent_based/inference/optimization 
Phase 1 (top-level reorg):
  - matrix_riccati promoted to crate root via re-export
  - new top-level groups: bsde, pde, stochastic_control,
    agent_based, inference, optimization

Phase 2 (bsde):
  - theta_scheme: linear-BSDE theta-scheme
  - deep_bsde_bridge: ConditionalExpectation trait + driver

Phase 3 (pde):
  - fokker_planck: 1D forward FP with conservative central FD
  - hjb_multid: explicit n-D HJB on Cartesian grid (d <= 3)
  - elliptic_fd: 2D Poisson SOR solver

Phase 4 (stochastic_control):
  - optimal_switching: Snell envelope backward induction
  - pontryagin: 1D LQR Riccati shooting
  - two_sided_intensity_control: bilateral intensity control

Phase 5/6 (controls):
  - optimal_control::quadratic_impact_control (closed-form Riccati)
  - stochastic_control::two_sided_intensity_control

Phase 7 (mean_field + agent_based):
  - mean_field::mckean_vlasov: interacting-particle Euler scheme
  - agent_based::mod: generic interacting-agent simulator

Phase 8 (inference + optimization):
  - inference::robust_drift: Huber IRLS drift estimator
  - optimization::generative_calibration_hooks: GenerativeSampler trait
    + Gaussian MMD + finite-diff calibration step

Tests: 38 NEW tests, all passing (165/170 lib total; the 5 pre-existing
failures predate v1.1 and are tracked separately).

Versions bumped: Cargo 2.0.0-alpha.1, pyproject 2.0.0a1.

Deferred to subsequent v2.0.x bumps (parallelisable follow-ups):
PyO3 bindings, executed companion notebooks, Sphinx RST pages,
hfthot-lab-instance propagation.
 2026-05-12 12:02:07 +02:00
ThotDjehuty d780ed81d7 release(v1.1.0): additive CPU-only generic numerical primitives 
Adds 9 new top-level / sub-modules to the Rust API only (no Python
bindings yet), with at least one analytic unit test per module.

New Rust modules:
- optimal_control::matrix_riccati  (RK4 backward solver)
- timeseries_utils::nonsync_covariance  (Hayashi-Yoshida)
- timeseries_utils::wavelet  (Haar / Daubechies DWT and MODWT)
- risk_measures  (VaR, CVaR, projected sub-gradient CVaR minimisation)
- graph::laplacian + graph::spectral_clustering  (Jacobi + k-means++)
- topology  (Vietoris-Rips persistent homology, bottleneck distance)
- volterra  (Caputo Adams, Markovian lift, second-kind Volterra,
             Fourier inversion of characteristic functions)
- signatures  (truncated tensor signature, log-sig, random reservoir,
               Salvi-Cass-Lyons signature kernel, shuffle product)

All previously stable APIs untouched; abi3-py38 ABI preserved.
New module tests: 29/29 passing. Pre-existing 5 unrelated failures
unchanged.
 2026-05-12 10:59:09 +02:00