freealg 0.7.4__tar.gz → 0.7.6__tar.gz

{freealg-0.7.4 → freealg-0.7.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: freealg
-Version: 0.7.4
+Version: 0.7.6
 Summary: Free probability for large matrices
 Home-page: https://github.com/ameli/freealg
 Download-URL: https://github.com/ameli/freealg/archive/main.zip

freealg-0.7.6/freealg/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.7.6"

freealg-0.7.6/freealg/_algebraic_form/_constraints.py

@@ -0,0 +1,98 @@
+
+# SPDX-FileCopyrightText: Copyright 2025, Siavash Ameli <sameli@berkeley.edu>
+# SPDX-License-Identifier: BSD-3-Clause
+# SPDX-FileType: SOURCE
+#
+# This program is free software: you can redistribute it and/or modify it under
+# the terms of the license found in the LICENSE.txt file in the root directory
+# of this source tree.
+
+
+# =======
+# Imports
+# =======
+
+import numpy
+
+__all__ = ['build_moment_constraint_matrix']
+
+
+# ==========
+# series mul
+# ==========
+
+def _series_mul(a, b, q_max):
+
+    na = min(len(a), q_max + 1)
+    nb = min(len(b), q_max + 1)
+    out = numpy.zeros(q_max + 1, dtype=float)
+    for i in range(na):
+        if a[i] == 0.0:
+            continue
+        j_max = min(nb - 1, q_max - i)
+        if j_max >= 0:
+            out[i:i + j_max + 1] += a[i] * b[:j_max + 1]
+    return out
+
+
+# ==========
+# series pow
+# ==========
+
+def _series_pow(mser, j, q_max):
+    if j == 0:
+        out = numpy.zeros(q_max + 1, dtype=float)
+        out[0] = 1.0
+        return out
+    out = mser.copy()
+    for _ in range(1, j):
+        out = _series_mul(out, mser, q_max)
+    return out
+
+
+# ===============================
+# build moment constraints matrix
+# ===============================
+
+def build_moment_constraint_matrix(pairs, deg_z, s, mu):
+
+    mu = numpy.asarray(mu, dtype=float).ravel()
+    if mu.size == 0:
+        return numpy.zeros((0, len(pairs)), dtype=float)
+
+    # m(z) = -sum_{p>=0} mu_p / z^{p+1}; t = 1/z so m(t) = -sum mu_p t^{p+1}
+    r = mu.size - 1
+    q_max = r
+
+    mser = numpy.zeros(q_max + 1, dtype=float)
+    for p in range(mu.size):
+        q = p + 1
+        if q <= q_max:
+            mser[q] = -float(mu[p])
+
+    # Precompute (m(t))^j coefficients up to t^{q_max}
+    mpow = []
+    for j in range(s + 1):
+        mpow.append(_series_pow(mser, j, q_max))
+
+    # Constraints: coeff of t^q in Q(t) := t^{deg_z} P(1/t, m(t)) must be 0
+    # Q(t) = sum_{i,j} c_{i,j} * t^{deg_z - i} * (m(t))^j
+    n_coef = len(pairs)
+    B = numpy.zeros((q_max + 1, n_coef), dtype=float)
+
+    for k, (i, j) in enumerate(pairs):
+        shift = deg_z - i
+        if shift < 0:
+            continue
+        mj = mpow[j]
+        for q in range(q_max + 1):
+            qq = q - shift
+            if 0 <= qq <= q_max:
+                B[q, k] = mj[qq]
+
+    # Drop all-zero rows (can happen if index-set can't support higher moments)
+    row_norm = numpy.linalg.norm(B, axis=1)
+    keep = row_norm > 0.0
+    B = B[keep, :]
+
+    return B

{freealg-0.7.4 → freealg-0.7.6}/freealg/_algebraic_form/_continuation_algebraic.py

@@ -13,10 +13,11 @@
 
 import numpy
 from .._geometric_form._continuation_genus0 import joukowski_z
+from ._constraints import build_moment_constraint_matrix
 
 __all__ = ['sample_z_joukowski', 'filter_z_away_from_cuts', 'powers',
-           'fit_polynomial_relation', 'eval_P', 'eval_roots',
-           'build_sheets_from_roots']
+           'fit_polynomial_relation', 'sanity_check_stieltjes_branch',
+           'eval_P', 'eval_roots', 'build_sheets_from_roots']
 
 
 # ======================
@@ -131,7 +132,11 @@ def powers(x, deg):
 # =======================
 
 def fit_polynomial_relation(z, m, s, deg_z, ridge_lambda=0.0, weights=None,
-                            triangular=None, normalize=False):
+                            triangular=None, normalize=False,
+                            mu=None, mu_reg=None):
+    """
+    Fits polynomial P(z, m) = 0 with samples from the physical branch.
+    """
 
     z = numpy.asarray(z, dtype=complex).ravel()
     m = numpy.asarray(m, dtype=complex).ravel()
@@ -191,17 +196,88 @@ def fit_polynomial_relation(z, m, s, deg_z, ridge_lambda=0.0, weights=None,
     if w is not None:
         A = A * w[:, None]
 
-    s_col = numpy.max(numpy.abs(A), axis=0)
+    # Enforce real coefficients by solving: Re(A) c = 0 and Im(A) c = 0
+    Ar = numpy.vstack([A.real, A.imag])
+
+    s_col = numpy.max(numpy.abs(Ar), axis=0)
     s_col[s_col == 0.0] = 1.0
-    As = A / s_col[None, :]
+    As = Ar / s_col[None, :]
+
+    # Optional moment constraints B c = 0 (hard via nullspace, soft via
+    # weighted rows)
+    if mu is not None:
+        B = build_moment_constraint_matrix(pairs, deg_z, s, mu)
+        if B.shape[0] > 0:
+            Bs = B / s_col[None, :]
+
+            if mu_reg is None:
+                # Hard constraints: solve in nullspace of Bs
+                uB, sB, vhB = numpy.linalg.svd(Bs, full_matrices=True)
+                tolB = 1e-12 * (sB[0] if sB.size else 1.0)
+                rankB = int(numpy.sum(sB > tolB))
+                if rankB >= n_coef:
+                    raise RuntimeError(
+                        'Moment constraints leave no feasible coefficients.')
+
+                N = vhB[rankB:, :].T  # (n_coef, n_free)
+                AN = As @ N
+
+                if ridge_lambda > 0.0:
+                    L = numpy.sqrt(ridge_lambda) * numpy.eye(N.shape[1],
+                                                             dtype=float)
+                    AN = numpy.vstack([AN, L])
+
+                _, _, vhN = numpy.linalg.svd(AN, full_matrices=False)
+                y = vhN[-1, :]
+                coef_scaled = N @ y
+
+                coef = coef_scaled / s_col
+
+            else:
+                mu_reg = float(mu_reg)
+                if mu_reg > 0.0:
+                    As_aug = As
+                    Bs_w = numpy.sqrt(mu_reg) * Bs
+                    As_aug = numpy.vstack([As_aug, Bs_w])
+
+                    if ridge_lambda > 0.0:
+                        L = numpy.sqrt(ridge_lambda) * numpy.eye(n_coef,
+                                                                 dtype=float)
+                        As_aug = numpy.vstack([As_aug, L])
+
+                    _, _, vh = numpy.linalg.svd(As_aug, full_matrices=False)
+                    coef_scaled = vh[-1, :]
+                    coef = coef_scaled / s_col
+                else:
+                    # mu_reg == 0 => ignore constraints
+                    if ridge_lambda > 0.0:
+                        L = numpy.sqrt(ridge_lambda) * numpy.eye(n_coef,
+                                                                 dtype=float)
+                        As = numpy.vstack([As, L])
+
+                    _, _, vh = numpy.linalg.svd(As, full_matrices=False)
+                    coef_scaled = vh[-1, :]
+                    coef = coef_scaled / s_col
+
+        else:
+            # B has no effective rows -> proceed unconstrained
+            if ridge_lambda > 0.0:
+                L = numpy.sqrt(ridge_lambda) * numpy.eye(n_coef, dtype=float)
+                As = numpy.vstack([As, L])
+
+            _, _, vh = numpy.linalg.svd(As, full_matrices=False)
+            coef_scaled = vh[-1, :]
+            coef = coef_scaled / s_col
 
-    if ridge_lambda > 0.0:
-        L = numpy.sqrt(ridge_lambda) * numpy.eye(n_coef, dtype=complex)
-        As = numpy.vstack([As, L])
+    else:
+        # No moment constraints
+        if ridge_lambda > 0.0:
+            L = numpy.sqrt(ridge_lambda) * numpy.eye(n_coef, dtype=float)
+            As = numpy.vstack([As, L])
 
-    _, _, vh = numpy.linalg.svd(As, full_matrices=False)
-    coef_scaled = vh[-1, :]
-    coef = coef_scaled / s_col
+        _, _, vh = numpy.linalg.svd(As, full_matrices=False)
+        coef_scaled = vh[-1, :]
+        coef = coef_scaled / s_col
 
     full = numpy.zeros((deg_z + 1, s + 1), dtype=complex)
     for k, (i, j) in enumerate(pairs):
@@ -213,6 +289,77 @@ def fit_polynomial_relation(z, m, s, deg_z, ridge_lambda=0.0, weights=None,
     return full
 
 
+# =============================
+# sanity check stieltjes branch
+# =============================
+
+def sanity_check_stieltjes_branch(a_coeffs, x_min, x_max, eta=0.1,
+                                  n_x=64, y0=None, max_bad_frac=0.05):
+    """
+    Quick sanity check: does P(z,m)=0 admit a continuously trackable root with
+    Im(m)>0 along z=x+i*eta.
+    """
+
+    x_min = float(x_min)
+    x_max = float(x_max)
+    eta = float(eta)
+    n_x = int(n_x)
+    if n_x < 4:
+        n_x = 4
+
+    if y0 is None:
+        y0 = 10.0 * max(1.0, abs(x_min), abs(x_max))
+    y0 = float(y0)
+
+    z0 = 1j * y0
+    m0_target = -1.0 / z0
+
+    c0 = _poly_coef_in_m(numpy.array([z0]), a_coeffs)[0]
+    r0 = numpy.roots(c0[::-1])
+    if r0.size == 0:
+        return {'ok': False, 'frac_bad': 1.0, 'n_test': 0, 'n_bad': 0}
+
+    k0 = int(numpy.argmin(numpy.abs(r0 - m0_target)))
+    m_prev = r0[k0]
+
+    xs = numpy.linspace(x_min, x_max, n_x)
+    zs = xs + 1j * eta
+
+    n_bad = 0
+    n_ok = 0
+
+    for z in zs:
+        c = _poly_coef_in_m(numpy.array([z]), a_coeffs)[0]
+        r = numpy.roots(c[::-1])
+        if r.size == 0 or not numpy.all(numpy.isfinite(r)):
+            n_bad += 1
+            continue
+
+        k = int(numpy.argmin(numpy.abs(r - m_prev)))
+        m_sel = r[k]
+        m_prev = m_sel
+        n_ok += 1
+
+        if not numpy.isfinite(m_sel) or (m_sel.imag <= 0.0):
+            n_bad += 1
+
+    n_test = n_ok + (n_bad - (n_x - n_ok))
+    if n_test <= 0:
+        n_test = n_x
+
+    frac_bad = float(n_bad) / float(n_x)
+    ok = frac_bad <= float(max_bad_frac)
+
+    status = {
+        'ok': ok,
+        'frac_bad': frac_bad,
+        'n_test': n_x,
+        'n_bad': n_bad
+    }
+
+    return status
+
+
 # ======
 # eval P
 # ======
@@ -361,6 +508,10 @@ def eval_roots(z, a_coeffs):
 # =======================
 
 def track_one_sheet_on_grid(z, roots, sheet_seed, cuts=None, i0=None, j0=None):
+    """
+    This is mostly used for visualization of the sheets.
+    """
+
     z = numpy.asarray(z)
     n_y, n_x = z.shape
     s = roots.shape[1]
@@ -467,6 +618,7 @@ def track_one_sheet_on_grid(z, roots, sheet_seed, cuts=None, i0=None, j0=None):
 # =======================
 
 def build_sheets_from_roots(z, roots, m1, cuts=None, i0=None, j0=None):
+
     z = numpy.asarray(z)
     m1 = numpy.asarray(m1)
 

{freealg-0.7.4 → freealg-0.7.6}/freealg/_algebraic_form/_homotopy.py

@@ -62,7 +62,7 @@ def stieltjes_select_root(roots, z, w_prev=None):
 # stieltjes poly
 # ==============
 
-def stieltjes_poly(z, a, eps=None, height=1e4, steps=100):
+def stieltjes_poly(z, a, eps=None, height=1e+4, steps=100):
     """
     Evaluate the Stieltjes-branch solution m(z) of an algebraic equation.
 

{freealg-0.7.4 → freealg-0.7.6}/freealg/_algebraic_form/algebraic_form.py

@@ -11,12 +11,12 @@
 # Imports
 # =======
 
-import inspect
 import numpy
 from .._util import resolve_complex_dtype, compute_eig
 # from .._util import compute_eig
 from ._continuation_algebraic import sample_z_joukowski, \
-        filter_z_away_from_cuts, fit_polynomial_relation, eval_P
+        filter_z_away_from_cuts, fit_polynomial_relation, \
+        sanity_check_stieltjes_branch, eval_P
 from ._edge import evolve_edges, merge_edges
 from ._decompress import decompress_newton
 from ._decompress2 import decompress_coeffs
@@ -149,8 +149,7 @@ class AlgebraicForm(object):
         # Data type for complex arrays
         self.dtype = resolve_complex_dtype(dtype)
 
-        if inspect.isclass(A) and hasattr(A, "stieltjes") and \
-                callable(getattr(A, "stieltjes", None)):
+        if hasattr(A, 'stieltjes') and callable(getattr(A, 'stieltjes', None)):
             # This is one of the distribution objects, like MarchenkoPastur
             self.stieltjes = A.stieltjes
             self.n = 1
@@ -195,6 +194,7 @@ class AlgebraicForm(object):
 
         # Initialize
         self.a_coeffs = None               # Polynomial coefficients
+        self.status = None                 # Fitting status
         self.cache = {}                    # Cache inner-computations
 
     # ===
@@ -208,14 +208,42 @@ class AlgebraicForm(object):
             y_eps=2e-2,
             x_pad=0.0,
             triangular=None,
+            mu=None,
+            mu_reg=None,
             normalize=False,
             verbose=False):
         """
-        Fits polynomial.
+        Fit polynomial.
+
+        Parameters
+        ----------
+
+        deg_m : int
+            Degree :math:`\\deg_m(P)`
+
+        deg_z : int
+            Degree :math:`\\deg_z(P)`
+
+        mu : array_like, default=None
+            If an array :math:`[\\mu_0, \\mu_`, \\dots, \\mu_r]` is given,
+            it enforces the first :math:`r+1` moments. Note that :math:`\\mu_0`
+            should be :math:`1` to ensure unit mass. See also ``mu_reg`.
+
+        mu_reg: float, default=None
+            If `None`, the constraints ``mu`` are applied as hard constraint.
+            If a positive number, the constraints are applied as a soft
+            constraints with regularisation ``mu_reg``.
+
+        Notes
+        -----
+
+        When the input data are from an exact model, hard moment constraint is
+        preferred over soft constraint as the latter can hurt an already a good
+        fit.
         """
 
         # Very important: reset cache whenever this function is called. This
-        # also empties all references holdign a cache copy.
+        # also empties all references holding a cache copy.
         # self.cache.clear()
 
         z_fits = []
@@ -232,33 +260,56 @@ class AlgebraicForm(object):
         z_fit = filter_z_away_from_cuts(z_fit, self.support, y_eps=y_eps,
                                         x_pad=x_pad)
 
+        # Fitting (w_inf = None means adaptive weight selection)
         m1_fit = self.stieltjes(z_fit)
         a_coeffs = fit_polynomial_relation(z_fit, m1_fit, s=deg_m, deg_z=deg_z,
                                            ridge_lambda=reg,
                                            triangular=triangular,
-                                           normalize=normalize)
+                                           normalize=normalize, mu=mu,
+                                           mu_reg=mu_reg)
 
         self.a_coeffs = a_coeffs
 
+        # Reporting error
+        P_res = numpy.abs(eval_P(z_fit, m1_fit, a_coeffs))
+        res_max = numpy.max(P_res[numpy.isfinite(P_res)])
+        res_99_9 = numpy.quantile(P_res[numpy.isfinite(P_res)], 0.999)
+
+        # Check polynomial has Stieltjes root
+        x_min = self.lam_m - 1.0
+        x_max = self.lam_p + 1.0
+        status = sanity_check_stieltjes_branch(a_coeffs, x_min, x_max,
+                                               eta=max(y_eps, 1e-2), n_x=128,
+                                               max_bad_frac=0.05)
+
+        status['res_max'] = float(res_max)
+        status['res_99_9'] = float(res_99_9)
+        self.status = status
+
         if verbose:
-            P_res = numpy.abs(eval_P(z_fit, m1_fit, a_coeffs))
-            print("fit residual max:", numpy.max(P_res[numpy.isfinite(P_res)]))
-            print("fit residual 99.9%:",
-                  numpy.quantile(P_res[numpy.isfinite(P_res)], 0.999))
+            print(f'fit residual max  : {res_max:>0.4e}')
+            print(f'fit residual 99.9%: {res_99_9:>0.4e}')
 
-            print('\nCoefficients')
-            with numpy.printoptions(precision=4, suppress=True):
+            print('\nCoefficients (real)')
+            with numpy.printoptions(precision=8, suppress=True):
                 for i in range(a_coeffs.shape[0]):
                     for j in range(a_coeffs.shape[1]):
                         v = a_coeffs[i, j]
-                        print(f"{v.real:>+0.4f}{v.imag:>+0.4f}j", end=" ")
+                        print(f'{v.real:>+0.8f}', end=' ')
                     print('')
 
-            print('\nCoefficient Magnitudes')
-            with numpy.printoptions(precision=6, suppress=True):
-                print(numpy.abs(a_coeffs))
+            a_coeffs_img_norm = numpy.linalg.norm(a_coeffs.imag, ord='fro')
+            print(f'\nCoefficients (imag) norm: {a_coeffs_img_norm:>0.4e}')
 
-        return a_coeffs
+            if not status['ok']:
+                print("\nWARNING: sanity check failed:\n" +
+                      f"\tfrac_bad: {status['frac_bad']:>0.3f}\n" +
+                      f"\tn_bad   : {status['n_bad']}\n" +
+                      f"\tn_test  : {status['n_test']}")
+            else:
+                print('\nStieltjes sanity check: OK')
+
+        return a_coeffs, status
 
     # =============
     # generate grid
@@ -638,11 +689,16 @@ class AlgebraicForm(object):
             z_query = x + 1j * self.delta
 
             # Initial condition at t=0 (physical branch)
-            w0_list = self.stieltjes(z_query)
+            # w0_list = self.stieltjes(z_query)
+            stieltjes = numpy.vectorize(m_fn)
+            w0_list = stieltjes(z_query)
 
             # Times
             t = numpy.log(alpha)
 
+            # Ensure it starts from t = 0
+            t = numpy.concatenate([numpy.zeros(1), t])
+
             # Evolve
             W, ok = decompress_newton(
                 z_query, t, self.a_coeffs,
@@ -650,6 +706,9 @@ class AlgebraicForm(object):
 
             rho = W.imag / numpy.pi
 
+            # Remove time zero
+            rho = rho[1:, :]
+
             if verbose:
                 print("success rate per t:", ok.mean(axis=1))
 

{freealg-0.7.4 → freealg-0.7.6}/freealg/_util.py

@@ -14,7 +14,7 @@
 import numpy
 import scipy
 
-__all__ = ['resolve_complex_dtype', 'compute_eig']
+__all__ = ['resolve_complex_dtype', 'compute_eig', 'subsample_matrix']
 
 
 # =====================
@@ -70,3 +70,26 @@ def compute_eig(A, lower=False):
     eig = scipy.linalg.eigvalsh(A, lower=lower, driver='ev')
 
     return eig
+
+
+# ================
+# subsample matrix
+# ================
+
+def subsample_matrix(matrix, submatrix_size, seed=None):
+    """
+    Generate a random subsample of a larger matrix
+    """
+
+    if matrix.shape[0] != matrix.shape[1]:
+        raise ValueError("Matrix must be square")
+
+    n = matrix.shape[0]
+    if submatrix_size > n:
+        raise ValueError("Submatrix size cannot exceed matrix size")
+
+    rng = numpy.random.default_rng(seed)
+    idx = rng.choice(n, size=submatrix_size, replace=False)
+    idx = numpy.sort(idx)  # optional, preserves original ordering
+
+    return matrix[numpy.ix_(idx, idx)]

{freealg-0.7.4 → freealg-0.7.6}/freealg.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: freealg
-Version: 0.7.4
+Version: 0.7.6
 Summary: Free probability for large matrices
 Home-page: https://github.com/ameli/freealg
 Download-URL: https://github.com/ameli/freealg/archive/main.zip

{freealg-0.7.4 → freealg-0.7.6}/freealg.egg-info/SOURCES.txt

@@ -17,6 +17,7 @@ freealg.egg-info/not-zip-safe
 freealg.egg-info/requires.txt
 freealg.egg-info/top_level.txt
 freealg/_algebraic_form/__init__.py
+freealg/_algebraic_form/_constraints.py
 freealg/_algebraic_form/_continuation_algebraic.py
 freealg/_algebraic_form/_decompress.py
 freealg/_algebraic_form/_decompress2.py

freealg-0.7.4/freealg/__version__.py

@@ -1 +0,0 @@
-__version__ = "0.7.4"