Adding Unit Tests for dl_eval.py (#854)

* Adding Unit Tests for dl_eval.py

This pull request introduces unit tests for the compute_correct and compute_accuracy functions. These tests utilize the pytest framework and incorporate "dummy tensors" to simulate various input conditions.

* 🎨 Format Python code with psf/black

* Added additional unit tests for edge cases

Added 0 accuracy and between 0 and 1 Accuracy Unit tests

* adding unit tests edge cases

* fixing indentation

fixing indentation

* added Pytest parameterized tests

added Pytest parameterized testing

* 🎨 Format Python code with psf/black

---------

Co-authored-by: yeaitsurya <yeaitsurya@users.noreply.github.com>

tests/test_dl_eval.py

@@ -0,0 +1,42 @@
+import torch
+import pytest
+import torch.nn as nn
+from backend.dl.dl_eval import compute_correct, compute_accuracy
+
+
+@pytest.mark.parametrize(
+    "predicted, actual, expected_correct",
+    [
+        # Test case: All correct predictions
+        (torch.tensor([[0.1, 0.2, 0.7], [0.3, 0.4, 0.3]]), torch.tensor([2, 1]), 2),
+        # Test case: Some correct predictions
+        (torch.tensor([[0.8, 0.1, 0.1], [0.2, 0.4, 0.4]]), torch.tensor([0, 0]), 1),
+        # Test case: No correct predictions
+        (torch.tensor([[0.9, 0.1, 0.0], [0.7, 0.2, 0.1]]), torch.tensor([2, 1]), 0),
+    ],
+)
+def test_compute_correct(predicted, actual, expected_correct):
+    # Compute the number of correct predictions
+    correct = compute_correct(predicted, actual)
+
+    # Check if the number of correct predictions is correct
+    assert correct == expected_correct
+
+
+@pytest.mark.parametrize(
+    "predicted, actual, expected_accuracy",
+    [
+        # Test case: Accuracy of 1
+        (torch.tensor([[0.1, 0.2, 0.7], [0.3, 0.4, 0.3]]), torch.tensor([2, 1]), 1),
+        # Test case: Accuracy between 0 and 1
+        (torch.tensor([[0.8, 0.1, 0.1], [0.2, 0.4, 0.4]]), torch.tensor([0, 0]), 0.5),
+        # Test case: Accuracy of 0
+        (torch.tensor([[0.9, 0.1, 0.0], [0.7, 0.2, 0.1]]), torch.tensor([2, 1]), 0),
+    ],
+)
+def test_compute_accuracy(predicted, actual, expected_accuracy):
+    # Compute the accuracy
+    accuracy = compute_accuracy(predicted, actual)
+
+    # Check if the accuracy is correct
+    assert accuracy == expected_accuracy