Generalize symmetry computation

.github/workflows/ci.yml

@@ -48,7 +48,10 @@ jobs:
 
       - name: mamba setup
         if: matrix.install-method == 'mamba'
-        uses: mamba-org/provision-with-micromamba@v14
+        uses: mamba-org/setup-micromamba@v1
+        with:
+          environment-file: environment.yml
+          cache-downloads: true
 
       - name: Python setup
         if: matrix.install-method == 'pip'
@@ -58,10 +61,12 @@ jobs:
           check-latest: true
 
       - name: Install dependencies
+        env:
+          PYTHON_VERSION: ${{ matrix.python-version }}
         run: |
           python --version
-          pip install pytest-cov restructuredtext-lint pytest-xdist 'coverage!=6.3.0'
-          pip install .[all]
+          pip install pytest-cov pytest-xdist 'coverage!=6.3.0'
+          pip install -e .[all]
           pip freeze
 
       - name: List installed package versions (conda)

docs/changes/168.optimization.rst

@@ -0,0 +1,3 @@
+- add keyword for half of the image
+- distinguish between tensor and array in get_ifft
+- fix micromamba installation

pyproject.toml

@@ -1,26 +1,3 @@
 [build-system]
-requires = ["setuptools>=61.0"]
+requires = ["setuptools>=64"]
 build-backend = "setuptools.build_meta"
-
-[project]
-name = "radionets"
-version = "0.2.0"
-authors = [
-  { name="Kevin Schmidt", email="[email protected]" },
-]
-description = "Imaging radio interferometric data with Neural Networks."
-readme = "README.md"
-requires-python = ">=3.7"
-classifiers = [
-    "Programming Language :: Python :: 3",
-    "License :: OSI Approved :: MIT License",
-    "Operating System :: OS Independent",
-    "Topic :: Scientific/Engineering :: Astronomy",
-    "Topic :: Scientific/Engineering :: Physics",
-    "Topic :: Scientific/Engineering :: Artificial Intelligence",
-    "Topic :: Scientific/Engineering :: Information Analysis",
-]
-
-[project.urls]
-"Homepage" = "https://github.com/pypa/sampleproject"
-"Bug Tracker" = "https://github.com/pypa/sampleproject/issues"

radionets/evaluation/train_inspection.py

@@ -111,7 +111,7 @@ def get_prediction(conf, mode="test"):
         images["pred"] = pred
         images["indices"] = indices
 
-    if images["pred"].shape[-1] == 128:
+    if images["pred"].shape[-2] < images["pred"].shape[-1]:
         images = apply_symmetry(images)
 
     return images

radionets/evaluation/utils.py

@@ -301,8 +301,6 @@ def get_images(test_ds, num_images, rand=False, indices=None):
 
         img_test = test_ds[indices][0]
         img_true = test_ds[indices][1]
-        img_test = img_test[:, :, :65, :]
-        img_true = img_true[:, :, :65, :]
         return img_test, img_true, indices
     else:
         mean = test_ds[indices][0]
@@ -356,7 +354,10 @@ def get_ifft(array, amp_phase=False, scale=False):
         image(s) in image space
     """
     if len(array.shape) == 3:
-        array = array.unsqueeze(0)
+        if hasattr(array, "numpy"):
+            array = array.unsqueeze(0)
+        else:
+            array = array[np.newaxis, :]
     if amp_phase:
         if scale:
             amp = 10 ** (10 * array[:, 0] - 10) - 1e-10
@@ -439,18 +440,19 @@ def symmetry(image, key):
         image = torch.tensor(image)
     if len(image.shape) == 3:
         image = image.view(1, image.shape[0], image.shape[1], image.shape[2])
-    upper_half = image[:, :, :64, :].clone()
+    half_image = image.shape[-1] // 2
+    upper_half = image[:, :, :half_image, :].clone()
     a = torch.rot90(upper_half, 2, dims=[-2, -1])
 
-    image[:, 0, 65:, 1:] = a[:, 0, :-1, :-1]
-    image[:, 0, 65:, 0] = a[:, 0, :-1, -1]
+    image[:, 0, half_image + 1 :, 1:] = a[:, 0, :-1, :-1]
+    image[:, 0, half_image + 1 :, 0] = a[:, 0, :-1, -1]
 
     if key == "unc":
-        image[:, 1, 65:, 1:] = a[:, 1, :-1, :-1]
-        image[:, 1, 65:, 0] = a[:, 1, :-1, -1]
+        image[:, 1, half_image + 1 :, 1:] = a[:, 1, :-1, :-1]
+        image[:, 1, half_image + 1 :, 0] = a[:, 1, :-1, -1]
     else:
-        image[:, 1, 65:, 1:] = -a[:, 1, :-1, :-1]
-        image[:, 1, 65:, 0] = -a[:, 1, :-1, -1]
+        image[:, 1, half_image + 1 :, 1:] = -a[:, 1, :-1, :-1]
+        image[:, 1, half_image + 1 :, 0] = -a[:, 1, :-1, -1]
 
     return image
 
@@ -473,8 +475,12 @@ def apply_symmetry(img_dict):
         if key != "indices":
             if isinstance(img_dict[key], np.ndarray):
                 img_dict[key] = torch.tensor(img_dict[key])
+            half_image = img_dict[key].shape[-1] // 2
             output = F.pad(
-                input=img_dict[key], pad=(0, 0, 0, 63), mode="constant", value=0
+                input=img_dict[key],
+                pad=(0, 0, 0, half_image - 1),
+                mode="constant",
+                value=0,
             )
             output = symmetry(output, key)
             img_dict[key] = output

setup.cfg

@@ -1,5 +1,60 @@
+[metadata]
+name = radionets
+version = 0.3.1
+author = Kevin Schmidt, Felix Geyer
+author_email = [email protected], [email protected]
+license = MIT
+description = Imaging radio interferometric data with neural networks
+url = https://github.com/radionets-project/radionets
+classifiers =
+    Development Status :: 4 - Beta
+    Intended Audience :: Science/Research
+    License :: OSI Approved :: MIT License
+    Natural Language :: English
+    Operating System :: OS Independent
+    Programming Language :: Python
+    Programming Language :: Python :: 3.6
+    Programming Language :: Python :: 3.7
+    Programming Language :: Python :: 3.8
+    Programming Language :: Python :: 3 :: Only
+    Topic :: Scientific/Engineering :: Astronomy
+    Topic :: Scientific/Engineering :: Physics
+    Topic :: Scientific/Engineering :: Artificial Intelligence
+    Topic :: Scientific/Engineering :: Information Analysis
+
 [aliases]
 test = pytest
 
+[options]
+packages = find:
+zip_safe = False
+setup_requires = pytest-runner
+install_requires =
+    fastai
+    kornia
+    pytorch-msssim
+    numpy
+    astropy
+    tqdm
+    click
+    numba
+    jupyter
+    h5py
+    scikit-image
+    pandas
+    toml
+    pytest
+    pytest-cov
+    pytest-order
+    comet_ml
+    pre-commit
+tests_require = pytest
+
 [tool:pytest]
-addopts = --verbose
+addopts = --verbose
+
+[options.entry_points]
+console_scripts =
+    radionets_simulations = radionets.simulations.scripts.simulate_images:main
+    radionets_training = radionets.dl_training.scripts.start_training:main
+    radionets_evaluation = radionets.evaluation.scripts.start_evaluation:main

tests/test_evaluation.py

@@ -67,6 +67,21 @@ def test_get_prediction(self):
         out_path.mkdir(parents=True, exist_ok=True)
         save_pred(str(out_path) + "/predictions_model_eval.h5", img)
 
+    def test_get_ifft(self):
+        import numpy as np
+        import torch
+
+        from radionets.evaluation.utils import get_ifft
+
+        a = torch.zeros([10, 2, 64, 64])
+        test_torch = get_ifft(a, amp_phase=True)
+        b = np.zeros([2, 64, 64])
+        test_numpy = get_ifft(b, amp_phase=True)
+        print(test_numpy.shape)
+        assert ~np.isnan([test_torch]).any()
+        assert ~np.isnan([test_numpy]).any()
+        assert len(test_torch.shape) == len(test_numpy.shape) + 1
+
     def test_contour(self):
         import numpy as np
         import toml
@@ -329,26 +344,11 @@ def test_gan_sources(self):
     def test_symmetry(self):
         import torch
 
-        from radionets.dl_framework.model import symmetry
-
-        x = torch.randint(0, 9, size=(1, 2, 4, 4))
-        x_symm = symmetry(x.clone())
-        for i in range(x.shape[-1]):
-            for j in range(x.shape[-1]):
-                assert (
-                    x_symm[0, 0, i, j]
-                    == x_symm[0, 0, x.shape[-1] - 1 - i, x.shape[-1] - 1 - j]
-                )
-                assert (
-                    x_symm[0, 1, i, j]
-                    == -x_symm[0, 1, x.shape[-1] - 1 - i, x.shape[-1] - 1 - j]
-                )
-
-        rot_amp = torch.rot90(x_symm[0, 0], 2)
-        rot_phase = torch.rot90(x_symm[0, 1], 2)
-
-        assert torch.isclose(rot_amp - x_symm[0, 0], torch.tensor(0)).all()
-        assert torch.isclose(rot_phase + x_symm[0, 1], torch.tensor(0)).all()
+        from radionets.evaluation.utils import symmetry
+
+        x = torch.randint(0, 9, size=(1, 2, 64, 64))
+        x_symm = symmetry(x.clone(), key="unc")
+        assert x_symm.shape == x.shape
 
     def test_sample_images(self):
         import numpy as np