invrs-opt 0.5.2__py3-none-any.whl → 0.7.0__py3-none-any.whl

invrs_opt/{lbfgsb/transform.py → parameterization/transforms.py}

@@ -62,6 +62,20 @@ def density_gaussian_filter_and_tanh(
     return transformed_density
 
 
+def _gaussian_kernel(fwhm: float, fwhm_size_multiple: float) -> jnp.ndarray:
+    """Returns a Gaussian kernel with the specified full-width at half-maximum."""
+    with jax.ensure_compile_time_eval():
+        kernel_size = max(1, int(jnp.ceil(fwhm * fwhm_size_multiple)))
+    # Ensure the kernel size is odd, so that there is always a central pixel which will
+    # contain the peak value of the Gaussian.
+    kernel_size += (kernel_size + 1) % 2
+    d = jnp.arange(0.5, kernel_size) - kernel_size / 2
+    x = d[:, jnp.newaxis]
+    y = d[jnp.newaxis, :]
+    sigma = fwhm / (2 * jnp.sqrt(2 * jnp.log(2)))
+    return jnp.exp(-(x**2 + y**2) / (2 * sigma**2))
+
+
 def normalized_array_from_density(density: types.Density2DArray) -> jnp.ndarray:
     """Returns an array with values scaled to the range `(-1, 1)`."""
     value_mid = (density.upper_bound + density.lower_bound) / 2
@@ -154,15 +168,66 @@ def _pad_mode_for_density(density: types.Density2DArray) -> Union[str, Tuple[str
     )
 
 
-def _gaussian_kernel(fwhm: float, fwhm_size_multiple: float) -> jnp.ndarray:
-    """Returns a Gaussian kernel with the specified full-width at half-maximum."""
+def resample(
+    x: jnp.ndarray,
+    shape: Tuple[int, ...],
+    method: jax.image.ResizeMethod = jax.image.ResizeMethod.LINEAR,
+) -> jnp.ndarray:
+    """Resamples `x` to have the specified `shape`.
+
+    The algorithm first upsamples `x` so that the pixels in the output image are
+    comprised of an integer number of pixels in the upsampled `x`, and then
+    performs box downsampling.
+
+    Args:
+        x: The array to be resampled.
+        shape: The shape of the output array.
+        method: The method used to resize `x` prior to box downsampling.
+
+    Returns:
+        The resampled array.
+    """
+    if x.ndim != len(shape):
+        raise ValueError(
+            f"`shape` must have length matching number of dimensions in `x`, "
+            f"but got {shape} when `x` had shape {x.shape}."
+        )
+
     with jax.ensure_compile_time_eval():
-        kernel_size = max(1, int(jnp.ceil(fwhm * fwhm_size_multiple)))
-    # Ensure the kernel size is odd, so that there is always a central pixel which will
-    # contain the peak value of the Gaussian.
-    kernel_size += (kernel_size + 1) % 2
-    d = jnp.arange(0.5, kernel_size) - kernel_size / 2
-    x = d[:, jnp.newaxis]
-    y = d[jnp.newaxis, :]
-    sigma = fwhm / (2 * jnp.sqrt(2 * jnp.log(2)))
-    return jnp.exp(-(x**2 + y**2) / (2 * sigma**2))
+        factor = [int(jnp.ceil(dx / d)) for dx, d in zip(x.shape, shape)]
+        upsampled_shape = tuple([d * f for d, f in zip(shape, factor)])
+
+    x_upsampled = jax.image.resize(
+        image=x,
+        shape=upsampled_shape,
+        method=method,
+    )
+
+    return box_downsample(x_upsampled, shape)
+
+
+def box_downsample(x: jnp.ndarray, shape: Tuple[int, ...]) -> jnp.ndarray:
+    """Downsamples `x` to a coarser resolution array using box downsampling.
+
+    Box downsampling forms nonoverlapping windows and simply averages the
+    pixels within each window. For example, downsampling `(0, 1, 2, 3, 4, 5)`
+    with a factor of `2` yields `(0.5, 2.5, 4.5)`.
+
+    Args:
+        x: The array to be downsampled.
+        shape: The shape of the output array; each axis dimension must evenly
+            divide the corresponding axis dimension in `x`.
+
+    Returns:
+        The output array with shape `shape`.
+    """
+    if x.ndim != len(shape) or any([(d % s) != 0 for d, s in zip(x.shape, shape)]):
+        raise ValueError(
+            f"Each axis of `shape` must evenly divide the corresponding axis "
+            f"dimension in `x`, but got shape {shape} when `x` has shape "
+            f"{x.shape}."
+        )
+    shape = sum([(s, d // s) for d, s in zip(x.shape, shape)], ())
+    axes = list(range(1, 2 * x.ndim, 2))
+    x = x.reshape(shape)
+    return jnp.mean(x, axis=axes)

{invrs_opt-0.5.2.dist-info → invrs_opt-0.7.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: invrs_opt
-Version: 0.5.2
+Version: 0.7.0
 Summary: Algorithms for inverse design
 Author-email: "Martin F. Schubert" <mfschubert@gmail.com>
 Maintainer-email: "Martin F. Schubert" <mfschubert@gmail.com>
@@ -34,22 +34,23 @@ Requires-Dist: jax
 Requires-Dist: jaxlib
 Requires-Dist: numpy
 Requires-Dist: requests
+Requires-Dist: optax
 Requires-Dist: scipy
 Requires-Dist: totypes
 Requires-Dist: types-requests
 Provides-Extra: dev
-Requires-Dist: bump-my-version ; extra == 'dev'
-Requires-Dist: darglint ; extra == 'dev'
-Requires-Dist: mypy ; extra == 'dev'
-Requires-Dist: pre-commit ; extra == 'dev'
+Requires-Dist: bump-my-version; extra == "dev"
+Requires-Dist: darglint; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: pre-commit; extra == "dev"
 Provides-Extra: tests
-Requires-Dist: parameterized ; extra == 'tests'
-Requires-Dist: pytest ; extra == 'tests'
-Requires-Dist: pytest-cov ; extra == 'tests'
-Requires-Dist: pytest-subtests ; extra == 'tests'
+Requires-Dist: parameterized; extra == "tests"
+Requires-Dist: pytest; extra == "tests"
+Requires-Dist: pytest-cov; extra == "tests"
+Requires-Dist: pytest-subtests; extra == "tests"
 
 # invrs-opt - Optimization algorithms for inverse design
-`v0.5.2`
+`v0.7.0`
 
 ## Overview
 

invrs_opt-0.7.0.dist-info/RECORD

@@ -0,0 +1,20 @@
+invrs_opt/__init__.py,sha256=cRGTL0pWiEmPY26dG0bfKaS_RV5I5N3FdFkKLM8Z-Wk,585
+invrs_opt/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+invrs_opt/experimental/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+invrs_opt/experimental/client.py,sha256=t4XxnditYbM9DWZeyBPj0Sa2acvkikT0ybhUdmH2r-Y,4852
+invrs_opt/experimental/labels.py,sha256=dQDAMPyFMV6mXnMy295z8Ap205DRdVzysXny_Be8FmY,562
+invrs_opt/optimizers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+invrs_opt/optimizers/base.py,sha256=ol9skiS4_itx6pmqTx49fqqRZGzHHSxjKPKguy6602s,1199
+invrs_opt/optimizers/lbfgsb.py,sha256=d7i02NZZ3yYdJg7wkERDMPpdBD3GaRPhmQexGrZPz_Y,34597
+invrs_opt/optimizers/wrapped_optax.py,sha256=L836gwzWbwxPNWh8Y7PSgFHV4PZluSklnbh3BR5djlc,11859
+invrs_opt/parameterization/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+invrs_opt/parameterization/base.py,sha256=QTnhOfMYbDchZOFzk9graryMd6rYHlyd7E2T1TtucB8,3570
+invrs_opt/parameterization/filter_project.py,sha256=XCPqQ2ECv7DDTLRtVGJePfnjKYB2XndI6DssSr-4MZw,3239
+invrs_opt/parameterization/gaussian_levelset.py,sha256=OYXPNC3GIGqCxea-u7rTXCOAEERLoict9kqnjYvGGto,24838
+invrs_opt/parameterization/pixel.py,sha256=4qCYDUCcFPr8W94whX5YFllrXgyZbPBVIJBf8m_Dv4k,1173
+invrs_opt/parameterization/transforms.py,sha256=8GzaIsUuuXvMCLiqAEEfxmi9qE9KqHzbuTj_m0GjH3w,8216
+invrs_opt-0.7.0.dist-info/LICENSE,sha256=ty6jHPvpyjHy6dbhnu6aDSY05bbl2jQTjnq9u1sBCfM,1078
+invrs_opt-0.7.0.dist-info/METADATA,sha256=EBEC3SgXD2hwdZ5s-z2dydbr5oSPSsgB34iMSdhaLeE,3339
+invrs_opt-0.7.0.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
+invrs_opt-0.7.0.dist-info/top_level.txt,sha256=hOziS2uJ_NgwaW9yhtOfeuYnm1X7vobPBcp_6eVWKfM,10
+invrs_opt-0.7.0.dist-info/RECORD,,

{invrs_opt-0.5.2.dist-info → invrs_opt-0.7.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.43.0)
+Generator: setuptools (72.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

invrs_opt-0.5.2.dist-info/RECORD

@@ -1,14 +0,0 @@
-invrs_opt/__init__.py,sha256=TPyc9kznbXNbM67XmXpnFIGTmGFQmYzPojKRueSZelU,309
-invrs_opt/base.py,sha256=dSX9QkMPzI8ROxy2cFNmMwk_89eQbk0rw94CzvLPQoY,907
-invrs_opt/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-invrs_opt/experimental/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-invrs_opt/experimental/client.py,sha256=td5o_YqqbcSypDrWCVrHGSoF8UxEdOLtKU0z9Dth9lA,4842
-invrs_opt/experimental/labels.py,sha256=dQDAMPyFMV6mXnMy295z8Ap205DRdVzysXny_Be8FmY,562
-invrs_opt/lbfgsb/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-invrs_opt/lbfgsb/lbfgsb.py,sha256=EKNTDTwEHcO_JeOIckMHVer5hxYcaNuqq9EZgnFnWJk,27820
-invrs_opt/lbfgsb/transform.py,sha256=a_Saj9Wq4lvqCJBrg5L2Z9DZ2NVs1xqrHLqha90a9Ws,5971
-invrs_opt-0.5.2.dist-info/LICENSE,sha256=ty6jHPvpyjHy6dbhnu6aDSY05bbl2jQTjnq9u1sBCfM,1078
-invrs_opt-0.5.2.dist-info/METADATA,sha256=SgIEqMR9ybcipS1NnHi-KSVTSZ3BzWKrv1ctw4jfqcE,3326
-invrs_opt-0.5.2.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-invrs_opt-0.5.2.dist-info/top_level.txt,sha256=hOziS2uJ_NgwaW9yhtOfeuYnm1X7vobPBcp_6eVWKfM,10
-invrs_opt-0.5.2.dist-info/RECORD,,