Gradicon codestyle fix

.github/workflows/cpu-test-action.yml

@@ -10,13 +10,15 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       max-parallel: 5
+      matrix:
+        python-version: [ '3.7', '3.8', '3.9', '3.11', '3.12']
 
     steps:
     - uses: actions/checkout@v2
-    - name: Set up Python 3.7
+    - name: Set up Python 
       uses: actions/setup-python@v2
       with:
-        python-version: 3.7
+        python-version: ${{ matrix.python-version }}
     - name: Install dependencies
       run: |
         pip install -e .

.gitignore

@@ -1,4 +1,5 @@
 results/
+evaluation_results/
 network_weights/
 __pycache__/
 *.swp

.readthedocs.yaml

@@ -13,7 +13,7 @@ build:
 
   jobs:
     post_install:
-      - pip install monai
+      - pip install monai sphinx-rtd-theme
 # Build documentation in the docs/ directory with Sphinx
 sphinx:
    configuration: docs/source/conf.py

README.md

@@ -16,7 +16,7 @@ _ICCV 2021_ https://arxiv.org/abs/2105.04459
 
 **GradICON: Approximate Diffeomorphisms via Gradient Inverse Consistency.**  
 Lin Tian, Hastings Greer, Francois-Xavier Vialard, Roland Kwitt, Raúl San José Estépar, Marc Niethammer.  
-_Arxiv preprint_ https://arxiv.org/abs/2206.05897
+_CVPR 2023_ https://arxiv.org/abs/2206.05897
 
 ## Cite this work
 ```
@@ -29,6 +29,7 @@ _Arxiv preprint_ https://arxiv.org/abs/2206.05897
     pages     = {3396-3405}
 }
 ```
+
 ```
 @article{Tian_2022_arXiv,
   title={GradICON: Approximate Diffeomorphisms via Gradient Inverse Consistency},

docs/source/conf.py

@@ -47,8 +47,11 @@
 
 
 def line_number(info):
+    # linkcode gets angry about namedtuple lol
     if "ICONLoss" in info["fullname"]:
         return 22
+    if "BendingLoss" in info["fullname"]:
+        return 235
     mod = icon_registration
     for elem in info["module"].split(".")[1:]:
         mod = getattr(mod, elem)

docs/source/icon_registration.pretrained_models.rst

@@ -18,3 +18,11 @@ icon\_registration.pretrained\_models.lung_ct
    :undoc-members:
    :show-inheritance:
 
+icon\_registration.pretrained\_models.HCP_brain
+-------------------------------------------------------
+
+.. automodule:: icon_registration.pretrained_models.HCP_brain
+   :members:
+   :undoc-members:
+   :show-inheritance:
+

docs/source/register_fives.rst

@@ -1,6 +1,8 @@
 Get Started: Register MNIST 5s
 ==============================
 
+The code in this tutorial is available as a `colab notebook <https://colab.research.google.com/drive/1Tu1w4DFEVUQM7wON4rY-nx97KTUq8ttR?usp=sharing>`_
+
 To learn to use ICON, we will train a model that registers pictures of the number 5 from the MNIST dataset.
 
 First create and activate some sort of virtual environment, and install torch into it `going by the official instructions <https://pytorch.org/get-started/locally/>`_
@@ -32,6 +34,7 @@ Next, create a git repo and import the code we need from `icon`.
    import numpy as np
    import torch
    import torchvision.utils
+   import matplotlib.pyplot as plt
 
 
 The function :func:`icon_registration.data.get_dataset_mnist` will give as a dataset of digits from MNIST filtered to only one number: this is useful for registration experiments, since there is a correct semantic map between two pictures of 5, but not between 5 and 8.