fixes

Makefile

@@ -12,13 +12,12 @@ pre-commit:
 	pre-commit install
 	pre-commit run --all
 
-tests:
-	@echo "Running unit tests..."
-	pytest ./lite/tests/text_generation -v
-
-integration-tests:
-	@echo "Running external integration tests..."
-	pytest ./lite/tests/text_generation -v
+test:
+	@echo "Running tests..."
+	pytest tests/classification
+	pytest tests/object_detection
+	pytest tests/semantic_segmentation
+	pytest tests/text_generation
 
 clean:
 	@echo "Cleaning up temporary files..."

docs/contributing.md

@@ -61,12 +61,7 @@ For linting and code formatting use:
 make pre-commit
 ```
 
-For unit and functional testing:
+Run all tests using:
 ```shell
 make tests
-```
-
-For integration testing:
-```shell
-make external-tests
 ```

src/valor_lite/classification/computation.py

@@ -291,7 +291,7 @@ def compute_confusion_matrix(
     confusion_matrix = np.full(
         (n_scores, n_labels, n_labels, 2 * n_examples + 1),
         fill_value=-1.0,
-        dtype=np.float64,
+        dtype=np.float32,
     )
     unmatched_ground_truths = np.full(
         (n_scores, n_labels, n_examples + 1),
@@ -389,4 +389,4 @@ def compute_confusion_matrix(
                         1 : misprd_label_examples.shape[0] + 1,
                     ] = misprd_label_examples[:, 0].flatten()
 
-    return confusion_matrix, unmatched_ground_truths
+    return confusion_matrix, unmatched_ground_truths  # type: ignore[reportReturnType]

src/valor_lite/object_detection/computation.py

@@ -685,12 +685,12 @@ def compute_confusion_matrix(
     confusion_matrix = -1 * np.ones(
         # (datum idx, gt idx, pd idx, pd score) * n_examples + count
         (n_ious, n_scores, n_labels, n_labels, 4 * n_examples + 1),
-        dtype=np.float64,
+        dtype=np.float32,
     )
     unmatched_predictions = -1 * np.ones(
         # (datum idx, pd idx, pd score) * n_examples + count
         (n_ious, n_scores, n_labels, 3 * n_examples + 1),
-        dtype=np.float64,
+        dtype=np.float32,
     )
     unmatched_ground_truths = -1 * np.ones(
         # (datum idx, gt idx) * n_examples + count