Merge branch 'main' of https://github.com/arnab39/EquivariantAdaptation

.coveragerc

@@ -1,12 +1,12 @@
 # .coveragerc to control coverage.py
 [run]
 branch = True
-source = equiadapt
+source = /src/equiadapt/
 # omit = bad_file.py
 
 [paths]
 source =
-    equiadapt/
+    /src/equiadapt/
     */site-packages/
 
 [report]

.pre-commit-config.yaml

@@ -56,8 +56,7 @@ repos:
   rev: 7.0.0
   hooks:
   - id: flake8
-  ## You can add flake8 plugins via `additional_dependencies`:
-  #  additional_dependencies: [flake8-bugbear]
+    additional_dependencies: [flake8-docstrings]
 
 ## Check for misspells in documentation files:
 # - repo: https://github.com/codespell-project/codespell

docs/conf.py

@@ -176,10 +176,7 @@
 # Theme options are theme-specific and customize the look and feel of a theme
 # further.  For a list of options available for each theme, see the
 # documentation.
-html_theme_options = {
-    "sidebar_width": "300px",
-    "page_width": "1200px"
-}
+html_theme_options = {"sidebar_width": "300px", "page_width": "1200px"}
 
 # Add any paths that contain custom themes here, relative to this directory.
 # html_theme_path = []
@@ -301,4 +298,4 @@
     "pyscaffold": ("https://pyscaffold.org/en/stable", None),
 }
 
-print(f"loading configurations for {project} {version} ...", file=sys.stderr)
+print(f"loading configurations for {project} {version} ...", file=sys.stderr)

equiadapt/images/canonicalization/continuous_group.py

@@ -1,6 +1,6 @@
 import math
 from omegaconf import DictConfig
-from typing import Any, List, Tuple, Union, Optional, Dict
+from typing import Optional, Dict, Any, Union, Tuple, List
 
 import kornia as K
 import torch
@@ -250,9 +250,10 @@ def get_groupelement(self, x: torch.Tensor) -> dict:
         if not hasattr(self, "canonicalization_info_dict"):
             self.canonicalization_info_dict = {}
 
-        group_element_dict, group_element_representation = (
-            self.get_group_from_out_vectors(out_vectors)
-        )
+        (
+            group_element_dict,
+            group_element_representation,
+        ) = self.get_group_from_out_vectors(out_vectors)
         self.canonicalization_info_dict["group_element_matrix_representation"] = (
             group_element_representation
         )
@@ -384,9 +385,10 @@ def get_groupelement(self, x: torch.Tensor) -> dict:
         if not hasattr(self, "canonicalization_info_dict"):
             self.canonicalization_info_dict = {}
 
-        group_element_dict, group_element_representations = (
-            self.get_group_from_out_vectors(out_vectors)
-        )
+        (
+            group_element_dict,
+            group_element_representations,
+        ) = self.get_group_from_out_vectors(out_vectors)
         # Store the matrix representation of the group element for regularization and identity metric
         self.canonicalization_info_dict["group_element_matrix_representation"] = (
             group_element_representations

equiadapt/pointcloud/canonicalization/continuous_group.py

@@ -3,6 +3,7 @@
 
 from omegaconf import DictConfig
 import torch
+
 from equiadapt.common.basecanonicalization import ContinuousGroupCanonicalization
 from equiadapt.common.utils import gram_schmidt
 from typing import Any, List, Tuple, Union, Optional

examples/images/segmentation/model_utils.py

@@ -89,7 +89,6 @@ def forward(self, images, targets):
 
 
 class SAMModel(nn.Module):
-
     def __init__(
         self,
         architecture_type: str,
@@ -148,7 +147,6 @@ def forward(self, images, targets):
 
 
 class FocalLoss(nn.Module):
-
     def __init__(self, weight=None, size_average=True):
         super().__init__()
         self.name = "focal_loss"
@@ -169,7 +167,6 @@ def forward(self, inputs, targets, alpha=ALPHA, gamma=GAMMA, smooth=1):
 
 
 class DiceLoss(nn.Module):
-
     def __init__(self, weight=None, size_average=True):
         super().__init__()
         self.name = "dice_loss"

examples/images/segmentation/prepare/coco_data.py

@@ -12,7 +12,6 @@
 
 
 class ResizeAndPad:
-
     def __init__(self, target_size):
         self.target_size = target_size
         self.transform = ResizeLongestSide(target_size)
@@ -126,7 +125,6 @@ def test_dataloader(self):
 
 
 class COCODataset(Dataset):
-
     def __init__(self, root_dir, annotation_file, transform=None, sam_transform=None):
         self.root_dir = root_dir
         self.transform = transform

setup.cfg

@@ -46,6 +46,7 @@ install_requires =
     torchvision
     kornia
     escnn @ git+https://github.com/danibene/escnn.git@remove/py3nj_dep
+    hydra-core
 
 [options.packages.find]
 exclude =
@@ -95,7 +96,7 @@ formats = bdist_wheel
 [flake8]
 # Some sane defaults for the code style checker flake8
 max_line_length = 130
-extend_ignore = E203, W503, E401, E501, E741, E266
+extend_ignore = E203, W503, E401, E501, E741, E266, D100, D107, D400, D401
 # ^  Black-compatible
 #    E203 and W503 have edge cases handled by black
 exclude =

tests/images/canonicalization/test_continuous_group.py

@@ -0,0 +1,75 @@
+from unittest.mock import Mock
+
+import pytest
+import torch
+from omegaconf import DictConfig
+
+from equiadapt import (
+    ContinuousGroupImageCanonicalization,
+)  # Update with your actual import path
+
+
+@pytest.fixture
+def sample_input():
+    # Create a sample input tensor
+    return torch.rand((1, 3, 64, 64))  # A batch with one color image of size 64x64
+
+
+@pytest.fixture
+def grayscale_input():
+    # Create a grayscale input tensor
+    return torch.rand((1, 1, 64, 64))  # A batch with one grayscale image of size 64x64
+
+
+@pytest.fixture
+def init_args():
+    # Mock initialization arguments
+    canonicalization_hyperparams = DictConfig(
+        {
+            "input_crop_ratio": 0.9,
+            "resize_shape": (32, 32),
+        }
+    )
+    return {
+        "canonicalization_network": torch.nn.Identity(),  # Placeholder
+        "canonicalization_hyperparams": canonicalization_hyperparams,
+        "in_shape": (3, 64, 64),
+    }
+
+
+def test_initialization(init_args):
+    cgic = ContinuousGroupImageCanonicalization(**init_args)
+    assert cgic.pad is not None, "Pad should be initialized."
+    assert cgic.crop is not None, "Crop should be initialized."
+
+
+def test_transformation_before_canonicalization_network_forward(
+    sample_input, init_args
+):
+    cgic = ContinuousGroupImageCanonicalization(**init_args)
+    transformed = cgic.transformations_before_canonicalization_network_forward(
+        sample_input
+    )
+    assert transformed.size() == torch.Size(
+        [1, 3, 32, 32]
+    ), "The transformed image should be resized to (32, 32)."
+
+
+@pytest.fixture
+def canonicalization_instance():
+    instance = ContinuousGroupImageCanonicalization(
+        canonicalization_network=Mock(),
+        canonicalization_hyperparams={
+            "input_crop_ratio": 0.9,
+            "resize_shape": (32, 32),
+        },
+        in_shape=(3, 64, 64),
+    )
+    # Mocking the get_groupelement method to return a fixed group element
+    instance.get_groupelement = Mock(
+        return_value={
+            "rotation": torch.eye(2).unsqueeze(0),
+            "reflection": torch.tensor([[[[0]]]]),
+        }
+    )
+    return instance

tox.ini

@@ -19,6 +19,7 @@ extras =
     testing
 commands =
     pytest {posargs}
+usedevelop=True
 
 
 # # To run `tox -e lint` you need to make sure you have a