Merge branch 'master' of https://github.com/cedricvincentcuaz/POT into partialfgw

Author: cedricvincentcuaz

RELEASES.md

@@ -41,6 +41,7 @@ This release also contains few bug fixes, concerning the support of any metric i
 - Notes before depreciating partial Gromov-Wasserstein function in `ot.partial` moved to ot.gromov  (PR #663)
 - Create `ot.gromov._partial` add new features `loss_fun = "kl_loss"` and `symmetry=False` to all solvers while increasing speed + updating adequatly `ot.solvers` (PR #663)
 - Added `ot.unbalanced.sinkhorn_unbalanced_translation_invariant` (PR #676)
+- Refactored `ot.bregman._convolutional` to improve readability (PR #709)
 
 #### Closed issues
 - Fixed `ot.gaussian` ignoring weights when computing means (PR #649, Issue #648)

ot/bregman/_convolutional.py

@@ -10,8 +10,53 @@
 
 import warnings
 
-from ..utils import list_to_array
 from ..backend import get_backend
+from ..utils import list_to_array
+
+_warning_msg = (
+    "Convolutional Sinkhorn did not converge. "
+    "Try a larger number of iterations `numItermax` "
+    "or a larger entropy `reg`."
+)
+
+
+def _get_convol_img_fn(nx, width, height, reg, type_as, log_domain=False):
+    """Return the convolution operator for 2D images.
+
+    The function constructed is equivalent to blurring on horizontal then vertical directions."""
+    t1 = nx.linspace(0, 1, width, type_as=type_as)
+    Y1, X1 = nx.meshgrid(t1, t1)
+    M1 = -((X1 - Y1) ** 2) / reg
+
+    t2 = nx.linspace(0, 1, height, type_as=type_as)
+    Y2, X2 = nx.meshgrid(t2, t2)
+    M2 = -((X2 - Y2) ** 2) / reg
+
+    # If normal domain is selected, we can use M1 and M2 to compute the convolution
+    if not log_domain:
+        K1, K2 = nx.exp(M1), nx.exp(M2)
+
+        def convol_imgs(imgs):
+            kx = nx.einsum("...ij,kjl->kil", K1, imgs)
+            kxy = nx.einsum("...ij,klj->kli", K2, kx)
+            return kxy
+
+    # Else, we can use M1 and M2 to compute the convolution in log-domain
+    else:
+
+        def convol_imgs(log_imgs):
+            log_imgs = nx.logsumexp(M1[:, :, None] + log_imgs[None], axis=1)
+            log_imgs = nx.logsumexp(M2[:, :, None] + log_imgs.T[None], axis=1).T
+            return log_imgs
+
+    return convol_imgs
+
+
+def _print_report(ii, err):
+    """Print the report of the iteration."""
+    if ii % 200 == 0:
+        print("{:5s}|{:12s}".format("It.", "Err") + "\n" + "-" * 19)
+    print("{:5d}|{:8e}|".format(ii, err))
 
 
 def convolutional_barycenter2d(
@@ -133,37 +178,26 @@ def _convolutional_barycenter2d(
     """
 
     A = list_to_array(A)
+    n_hists, width, height = A.shape
 
     nx = get_backend(A)
 
     if weights is None:
-        weights = nx.ones((A.shape[0],), type_as=A) / A.shape[0]
+        weights = nx.ones((n_hists,), type_as=A) / n_hists
     else:
-        assert len(weights) == A.shape[0]
+        assert len(weights) == n_hists
 
     if log:
         log = {"err": []}
 
-    bar = nx.ones(A.shape[1:], type_as=A)
+    bar = nx.ones((width, height), type_as=A)
     bar /= nx.sum(bar)
     U = nx.ones(A.shape, type_as=A)
     V = nx.ones(A.shape, type_as=A)
     err = 1
 
     # build the convolution operator
-    # this is equivalent to blurring on horizontal then vertical directions
-    t = nx.linspace(0, 1, A.shape[1], type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    K1 = nx.exp(-((X - Y) ** 2) / reg)
-
-    t = nx.linspace(0, 1, A.shape[2], type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    K2 = nx.exp(-((X - Y) ** 2) / reg)
-
-    def convol_imgs(imgs):
-        kx = nx.einsum("...ij,kjl->kil", K1, imgs)
-        kxy = nx.einsum("...ij,klj->kli", K2, kx)
-        return kxy
+    convol_imgs = _get_convol_img_fn(nx, width, height, reg, type_as=A)
 
     KU = convol_imgs(U)
     for ii in range(numItermax):
@@ -177,24 +211,18 @@ def convol_imgs(imgs):
             # log and verbose print
             if log:
                 log["err"].append(err)
-
             if verbose:
-                if ii % 200 == 0:
-                    print("{:5s}|{:12s}".format("It.", "Err") + "\n" + "-" * 19)
-                print("{:5d}|{:8e}|".format(ii, err))
+                _print_report(ii, err)
             if err < stopThr:
                 break
 
     else:
         if warn:
-            warnings.warn(
-                "Convolutional Sinkhorn did not converge. "
-                "Try a larger number of iterations `numItermax` "
-                "or a larger entropy `reg`."
-            )
+            warnings.warn(_warning_msg)
     if log:
         log["niter"] = ii
         log["U"] = U
+        log["V"] = V
         return bar, log
     else:
         return bar
@@ -218,6 +246,8 @@ def _convolutional_barycenter2d_log(
     A = list_to_array(A)
 
     nx = get_backend(A)
+    # This error is raised because we are using mutable assignment in the line
+    #  `log_KU[k] = ...` which is not allowed in Jax and TF.
     if nx.__name__ in ("jax", "tf"):
         raise NotImplementedError(
             "Log-domain functions are not yet implemented"
@@ -236,19 +266,7 @@ def _convolutional_barycenter2d_log(
 
     err = 1
     # build the convolution operator
-    # this is equivalent to blurring on horizontal then vertical directions
-    t = nx.linspace(0, 1, width, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    M1 = -((X - Y) ** 2) / reg
-
-    t = nx.linspace(0, 1, height, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    M2 = -((X - Y) ** 2) / reg
-
-    def convol_img(log_img):
-        log_img = nx.logsumexp(M1[:, :, None] + log_img[None], axis=1)
-        log_img = nx.logsumexp(M2[:, :, None] + log_img.T[None], axis=1).T
-        return log_img
+    convol_img = _get_convol_img_fn(nx, width, height, reg, type_as=A, log_domain=True)
 
     logA = nx.log(A + stabThr)
     log_KU, G, F = nx.zeros((3, *logA.shape), type_as=A)
@@ -265,22 +283,15 @@ def convol_img(log_img):
             # log and verbose print
             if log:
                 log["err"].append(err)
-
             if verbose:
-                if ii % 200 == 0:
-                    print("{:5s}|{:12s}".format("It.", "Err") + "\n" + "-" * 19)
-                print("{:5d}|{:8e}|".format(ii, err))
+                _print_report(ii, err)
             if err < stopThr:
                 break
         G = log_bar[None, :, :] - log_KU
 
     else:
         if warn:
-            warnings.warn(
-                "Convolutional Sinkhorn did not converge. "
-                "Try a larger number of iterations `numItermax` "
-                "or a larger entropy `reg`."
-            )
+            warnings.warn(_warning_msg)
     if log:
         log["niter"] = ii
         return nx.exp(log_bar), log
@@ -417,23 +428,11 @@ def _convolutional_barycenter2d_debiased(
     bar /= width * height
     U = nx.ones(A.shape, type_as=A)
     V = nx.ones(A.shape, type_as=A)
-    c = nx.ones(A.shape[1:], type_as=A)
+    c = nx.ones((width, height), type_as=A)
     err = 1
 
     # build the convolution operator
-    # this is equivalent to blurring on horizontal then vertical directions
-    t = nx.linspace(0, 1, width, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    K1 = nx.exp(-((X - Y) ** 2) / reg)
-
-    t = nx.linspace(0, 1, height, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    K2 = nx.exp(-((X - Y) ** 2) / reg)
-
-    def convol_imgs(imgs):
-        kx = nx.einsum("...ij,kjl->kil", K1, imgs)
-        kxy = nx.einsum("...ij,klj->kli", K2, kx)
-        return kxy
+    convol_imgs = _get_convol_img_fn(nx, width, height, reg, type_as=A)
 
     KU = convol_imgs(U)
     for ii in range(numItermax):
@@ -451,26 +450,20 @@ def convol_imgs(imgs):
             # log and verbose print
             if log:
                 log["err"].append(err)
-
             if verbose:
-                if ii % 200 == 0:
-                    print("{:5s}|{:12s}".format("It.", "Err") + "\n" + "-" * 19)
-                print("{:5d}|{:8e}|".format(ii, err))
+                _print_report(ii, err)
 
             # debiased Sinkhorn does not converge monotonically
             # guarantee a few iterations are done before stopping
             if err < stopThr and ii > 20:
                 break
     else:
         if warn:
-            warnings.warn(
-                "Sinkhorn did not converge. You might want to "
-                "increase the number of iterations `numItermax` "
-                "or the regularization parameter `reg`."
-            )
+            warnings.warn(_warning_msg)
     if log:
         log["niter"] = ii
         log["U"] = U
+        log["V"] = V
         return bar, log
     else:
         return bar
@@ -492,6 +485,8 @@ def _convolutional_barycenter2d_debiased_log(
     A = list_to_array(A)
     n_hists, width, height = A.shape
     nx = get_backend(A)
+    # This error is raised because we are using mutable assignment in the line
+    #  `log_KU[k] = ...` which is not allowed in Jax and TF.
     if nx.__name__ in ("jax", "tf"):
         raise NotImplementedError(
             "Log-domain functions are not yet implemented"
@@ -507,19 +502,7 @@ def _convolutional_barycenter2d_debiased_log(
 
     err = 1
     # build the convolution operator
-    # this is equivalent to blurring on horizontal then vertical directions
-    t = nx.linspace(0, 1, width, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    M1 = -((X - Y) ** 2) / reg
-
-    t = nx.linspace(0, 1, height, type_as=A)
-    [Y, X] = nx.meshgrid(t, t)
-    M2 = -((X - Y) ** 2) / reg
-
-    def convol_img(log_img):
-        log_img = nx.logsumexp(M1[:, :, None] + log_img[None], axis=1)
-        log_img = nx.logsumexp(M2[:, :, None] + log_img.T[None], axis=1).T
-        return log_img
+    convol_img = _get_convol_img_fn(nx, width, height, reg, type_as=A, log_domain=True)
 
     logA = nx.log(A + stabThr)
     log_bar, c = nx.zeros((2, width, height), type_as=A)
@@ -540,22 +523,15 @@ def convol_img(log_img):
             # log and verbose print
             if log:
                 log["err"].append(err)
-
             if verbose:
-                if ii % 200 == 0:
-                    print("{:5s}|{:12s}".format("It.", "Err") + "\n" + "-" * 19)
-                print("{:5d}|{:8e}|".format(ii, err))
+                _print_report(ii, err)
             if err < stopThr and ii > 20:
                 break
         G = log_bar[None, :, :] - log_KU
 
     else:
         if warn:
-            warnings.warn(
-                "Convolutional Sinkhorn did not converge. "
-                "Try a larger number of iterations `numItermax` "
-                "or a larger entropy `reg`."
-            )
+            warnings.warn(_warning_msg)
     if log:
         log["niter"] = ii
         return nx.exp(log_bar), log