Format files

Author: c-salomonsen

utils/createfolders.py

@@ -1,6 +1,4 @@
-import argparse
 from pathlib import Path
-from tempfile import TemporaryDirectory
 
 
 def createfolders(*dirs: Path) -> None:

utils/dataloaders/uspsh5_7_9.py

@@ -1,9 +1,9 @@
-from torch.utils.data import Dataset
-import numpy as np
 import h5py
-from torchvision import transforms
-from PIL import Image
+import numpy as np
 import torch
+from PIL import Image
+from torch.utils.data import Dataset
+from torchvision import transforms
 
 
 class USPSH5_Digit_7_9_Dataset(Dataset):
@@ -95,14 +95,20 @@ def __getitem__(self, id):
 
 def main():
     # Example Usage:
-    transform = transforms.Compose([
-        transforms.Resize((16, 16)),  # Ensure images are 16x16
-        transforms.ToTensor(),
-        transforms.Normalize((0.5,), (0.5,))  # Normalize to [-1, 1]
-    ])
+    transform = transforms.Compose(
+        [
+            transforms.Resize((16, 16)),  # Ensure images are 16x16
+            transforms.ToTensor(),
+            transforms.Normalize((0.5,), (0.5,)),  # Normalize to [-1, 1]
+        ]
+    )
 
     # Load the dataset
-    dataset = USPSH5_Digit_7_9_Dataset(h5_path="C:/Users/Solveig/OneDrive/Dokumente/UiT PhD/Courses/Git/usps.h5", mode="train", transform=transform)
+    dataset = USPSH5_Digit_7_9_Dataset(
+        h5_path="C:/Users/Solveig/OneDrive/Dokumente/UiT PhD/Courses/Git/usps.h5",
+        mode="train",
+        transform=transform,
+    )
     data_loader = torch.utils.data.DataLoader(dataset, batch_size=2, shuffle=True)
     batch = next(iter(data_loader))  # grab a batch from the dataloader
     img, label = batch
@@ -112,5 +118,6 @@ def main():
     # Check dataset size
     print(f"Dataset size: {len(dataset)}")
 
-if __name__ == '__main__':
-    main()
+
+if __name__ == "__main__":
+    main()

utils/metrics/F1.py

@@ -1,40 +1,41 @@
-import torch.nn as nn
 import torch
+import torch.nn as nn
 
 
 class F1Score(nn.Module):
     """
-          F1 Score implementation with direct averaging inside the compute method.
+    F1 Score implementation with direct averaging inside the compute method.
+
+    Parameters
+    ----------
+    num_classes : int
+        Number of classes.
 
-          Parameters
-          ----------
-          num_classes : int
-              Number of classes.
+    Attributes
+    ----------
+    num_classes : int
+        The number of classes.
 
-          Attributes
-          ----------
-          num_classes : int
-              The number of classes.
+    tp : torch.Tensor
+        Tensor for True Positives (TP) for each class.
 
-          tp : torch.Tensor
-              Tensor for True Positives (TP) for each class.
+    fp : torch.Tensor
+        Tensor for False Positives (FP) for each class.
 
-          fp : torch.Tensor
-              Tensor for False Positives (FP) for each class.
+    fn : torch.Tensor
+        Tensor for False Negatives (FN) for each class.
+    """
 
-          fn : torch.Tensor
-              Tensor for False Negatives (FN) for each class.
-          """
     def __init__(self, num_classes):
         """
-             Initializes the F1Score object, setting up the necessary state variables.
+        Initializes the F1Score object, setting up the necessary state variables.
 
-             Parameters
-             ----------
-             num_classes : int
-                 The number of classes in the classification task.
+        Parameters
+        ----------
+        num_classes : int
+            The number of classes in the classification task.
 
-             """
+        """
 
         super().__init__()
 
@@ -47,16 +48,16 @@ def __init__(self, num_classes):
 
     def update(self, preds, target):
         """
-       Update the variables with predictions and true labels.
+        Update the variables with predictions and true labels.
 
-       Parameters
-       ----------
-       preds : torch.Tensor
-           Predicted logits (shape: [batch_size, num_classes]).
+        Parameters
+        ----------
+        preds : torch.Tensor
+            Predicted logits (shape: [batch_size, num_classes]).
 
-       target : torch.Tensor
-           True labels (shape: [batch_size]).
-       """
+        target : torch.Tensor
+            True labels (shape: [batch_size]).
+        """
         preds = torch.argmax(preds, dim=1)
 
         # Calculate True Positives (TP), False Positives (FP), and False Negatives (FN) per class
@@ -76,17 +77,20 @@ def compute(self):
         """
 
         # Compute F1 score based on the specified averaging method
-        f1_score = 2 * torch.sum(self.tp) / (2 * torch.sum(self.tp) + torch.sum(self.fp) + torch.sum(self.fn))
+        f1_score = (
+            2
+            * torch.sum(self.tp)
+            / (2 * torch.sum(self.tp) + torch.sum(self.fp) + torch.sum(self.fn))
+        )
 
         return f1_score
 
 
 def test_f1score():
     f1_metric = F1Score(num_classes=3)
-    preds = torch.tensor([[0.8, 0.1, 0.1],
-                          [0.2, 0.7, 0.1],
-                          [0.2, 0.3, 0.5],
-                          [0.1, 0.2, 0.7]])
+    preds = torch.tensor(
+        [[0.8, 0.1, 0.1], [0.2, 0.7, 0.1], [0.2, 0.3, 0.5], [0.1, 0.2, 0.7]]
+    )
 
     target = torch.tensor([0, 1, 0, 2])