pre-commits and symmetry layer (#126)

* Add pre-commits to repository

* Add warning to symmetry layer

* Add test for symmetry layer

* Delete unused import

* Add another symmetry test

* Delete old symmetry function and rename

* Update url

* Update README.md

.pre-commit-config.yaml

@@ -0,0 +1,10 @@
+repos:
+  - repo: https://github.com/psf/black
+    rev: 22.6.0
+    hooks:
+      - id: black-jupyter
+  - repo: https://github.com/PyCQA/flake8
+    rev: 4.0.1
+    hooks:
+      - id: flake8
+        args: [--max-line-length=88, "--extend-ignore=E203"]

README.md

@@ -20,7 +20,11 @@ You can create one by running the following command in this repository:
 $ conda env create -f environment.yml
 ```
 Depending on your `cuda` version you have to specify the `cudatoolkit` version used by `pytorch`. If you are working on machines
-with `cuda` versions < 10.2, please change the version number in the environment.yml file.
+with `cuda` versions < 10.2, please change the version number in the environment.yml file. Since the package `pre-commit` is used, you need to execute 
+```
+$ pre-commit install
+```
+after the installation.
 
 ## Usage
 

radionets/dl_framework/architectures/unc_archs.py

@@ -1,13 +1,10 @@
 import torch
 from torch import nn
 from radionets.dl_framework.model import (
-    Lambda,
-    symmetry,
     GeneralELU,
     LocallyConnected2d,
 )
 from radionets.dl_framework.architectures.res_exp import SRResNet
-from functools import partial
 
 
 class Uncertainty(nn.Module):
@@ -43,8 +40,6 @@ def __init__(self, img_size):
             )
         )
 
-        self.symmetry = Lambda(partial(symmetry, mode="real"))
-
         self.elu = GeneralELU(add=+(1 + 1e-7))
 
     def forward(self, x):

radionets/dl_framework/model.py

@@ -3,7 +3,7 @@
 import torch.nn.functional as F
 from torch.nn.modules.utils import _pair
 from pathlib import Path
-from math import sqrt, pi
+from math import pi
 
 
 class Lambda(nn.Module):
@@ -15,30 +15,6 @@ def forward(self, x):
         return self.func(x)
 
 
-def symmetry(x, mode="real"):
-    center = (x.shape[1]) // 2
-    u = torch.arange(center)
-    v = torch.arange(center)
-
-    diag1 = torch.arange(center, x.shape[1])
-    diag2 = torch.arange(center, x.shape[1])
-    diag_indices = torch.stack((diag1, diag2))
-    grid = torch.tril_indices(x.shape[1], x.shape[1], -1)
-
-    x_sym = torch.cat((grid[0].reshape(-1, 1), diag_indices[0].reshape(-1, 1)))
-    y_sym = torch.cat((grid[1].reshape(-1, 1), diag_indices[1].reshape(-1, 1)))
-    x = torch.rot90(x, 1, dims=(1, 2))
-    i = center + (center - x_sym)
-    j = center + (center - y_sym)
-    u = center - (center - x_sym)
-    v = center - (center - y_sym)
-    if mode == "real":
-        x[:, i, j] = x[:, u, v]
-    if mode == "imag":
-        x[:, i, j] = -x[:, u, v]
-    return torch.rot90(x, 3, dims=(1, 2))
-
-
 class GeneralRelu(nn.Module):
     def __init__(self, leak=None, sub=None, maxv=None):
         super().__init__()
@@ -75,8 +51,8 @@ def init_cnn_(m, f):
         f(m.weight, a=0.1)
         if getattr(m, "bias", None) is not None:
             m.bias.data.zero_()
-    for l in m.children():
-        init_cnn_(l, f)
+    for c in m.children():
+        init_cnn_(c, f)
 
 
 def init_cnn(m, uniform=False):
@@ -223,7 +199,7 @@ def __init__(
                 in_channels,
                 output_size[0],
                 output_size[1],
-                kernel_size ** 2,
+                kernel_size**2,
             )
         )
         if bias:
@@ -313,7 +289,9 @@ def _conv_block(self, ni, nf, stride):
         )
 
 
-def even_better_symmetry(x):
+def symmetry(x):
+    if x.shape[-1] % 2 != 0:
+        raise ValueError("The symmetry function only works for even image sizes.")
     upper_half = x[:, :, 0 : x.shape[2] // 2, :].clone()
     upper_left = upper_half[:, :, :, 0 : upper_half.shape[3] // 2].clone()
     upper_right = upper_half[:, :, :, upper_half.shape[3] // 2 :].clone()

setup.py

@@ -4,7 +4,7 @@
     name="radionets",
     version="0.1.14",
     description="Imaging radio interferometric data with neural networks",
-    url="https://github.com/Kevin2/radionets",
+    url="https://github.com/radionets-project/radionets",
     author="Kevin Schmidt, Felix Geyer",
     author_email="[email protected], [email protected]",
     license="MIT",
@@ -25,14 +25,16 @@
         "pytest",
         "pytest-cov",
         "pytest-order",
-        "comet_ml"
+        "comet_ml",
+        "pre-commit",
     ],
     setup_requires=["pytest-runner"],
     tests_require=["pytest"],
     zip_safe=False,
     entry_points={
         "console_scripts": [
-            "radionets_simulations = radionets.simulations.scripts.simulate_images:main",
+            "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

@@ -117,9 +117,9 @@ def test_im_to_array_value(self):
 
         x_coords, y_coords, value = im_to_array_value(image)
 
-        assert x_coords.shape == (2, 64 ** 2)
-        assert y_coords.shape == (2, 64 ** 2)
-        assert value.shape == (2, 64 ** 2)
+        assert x_coords.shape == (2, 64**2)
+        assert y_coords.shape == (2, 64**2)
+        assert value.shape == (2, 64**2)
 
     def test_bmul(self):
         import torch
@@ -323,6 +323,29 @@ def test_gan_sources(self):
 
         assert evaluate_gan_sources(conf) is None
 
+    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()
+
     def test_evaluation(self):
         import shutil
         import os