🧪 Add tests

src/anomalib/__init__.py

@@ -22,4 +22,3 @@ class TaskType(str, Enum):
     CLASSIFICATION = "classification"
     DETECTION = "detection"
     SEGMENTATION = "segmentation"
-    EXPLANATION = "explanation"

src/anomalib/callbacks/metrics.py

@@ -75,7 +75,7 @@ def setup(
         pixel_metric_names: list[str] | dict[str, dict[str, Any]]
         if self.pixel_metric_names is None:
             pixel_metric_names = []
-        elif self.task in {TaskType.CLASSIFICATION, TaskType.EXPLANATION}:
+        elif self.task in TaskType.CLASSIFICATION:
             pixel_metric_names = []
             logger.warning(
                 "Cannot perform pixel-level evaluation when task type is {self.task.value}. "

src/anomalib/data/base/dataset.py

@@ -170,7 +170,7 @@ def __getitem__(self, index: int) -> dict[str, str | torch.Tensor]:
         image = read_image(image_path, as_tensor=True)
         item = {"image_path": image_path, "label": label_index}
 
-        if self.task in {TaskType.CLASSIFICATION, TaskType.EXPLANATION}:
+        if self.task in TaskType.CLASSIFICATION:
             item["image"] = self.transform(image) if self.transform else image
         elif self.task in {TaskType.DETECTION, TaskType.SEGMENTATION}:
             # Only Anomalous (1) images have masks in anomaly datasets

src/anomalib/data/base/depth.py

@@ -48,7 +48,7 @@ def __getitem__(self, index: int) -> dict[str, str | torch.Tensor]:
         depth_image = to_tensor(read_depth_image(depth_path))
         item = {"image_path": image_path, "depth_path": depth_path, "label": label_index}
 
-        if self.task in {TaskType.CLASSIFICATION, TaskType.EXPLANATION}:
+        if self.task == TaskType.CLASSIFICATION:
             item["image"], item["depth_image"] = (
                 self.transform(image, depth_image) if self.transform else (image, depth_image)
             )

src/anomalib/deploy/inferencers/openvino_inferencer.py

@@ -278,7 +278,7 @@ def post_process(self, predictions: np.ndarray, metadata: dict | DictConfig | No
             pred_idx = pred_score >= metadata["image_threshold"]
             pred_label = LabelName.ABNORMAL if pred_idx else LabelName.NORMAL
 
-        if task in {TaskType.CLASSIFICATION, TaskType.EXPLANATION}:
+        if task == TaskType.CLASSIFICATION:
             _, pred_score = self._normalize(pred_scores=pred_score, metadata=metadata)
         elif task in {TaskType.SEGMENTATION, TaskType.DETECTION}:
             if "pixel_threshold" in metadata:

src/anomalib/engine/engine.py

@@ -322,7 +322,7 @@ def _setup_trainer(self, model: AnomalyModule) -> None:
             self._cache.update(model)
 
         # Setup anomalib callbacks to be used with the trainer
-        self._setup_anomalib_callbacks()
+        self._setup_anomalib_callbacks(model)
 
         # Temporarily set devices to 1 to avoid issues with multiple processes
         self._cache.args["devices"] = 1
@@ -405,7 +405,7 @@ def _setup_transform(
                 if not getattr(dataloader.dataset, "transform", None):
                     dataloader.dataset.transform = transform
 
-    def _setup_anomalib_callbacks(self) -> None:
+    def _setup_anomalib_callbacks(self, model: AnomalyModule) -> None:
         """Set up callbacks for the trainer."""
         _callbacks: list[Callback] = []
 
@@ -433,7 +433,9 @@ def _setup_anomalib_callbacks(self) -> None:
         _callbacks.append(_MetricsCallback(self.task, self.image_metric_names, self.pixel_metric_names))
 
         visualizer: BaseVisualizer
-        if self.task == TaskType.EXPLANATION:
+
+        # TODO(ashwinvaidya17): temporary  # noqa: TD003 ignoring as visualizer is getting a complete overhaul
+        if model.__class__.__name__ == "VlmAd":
             visualizer = ExplanationVisualizer()
         else:
             visualizer = ImageVisualizer(task=self.task, normalize=self.normalization == NormalizationMethod.NONE)

src/anomalib/models/image/vlm_ad/lightning_model.py

@@ -99,3 +99,31 @@ def configure_transforms(image_size: tuple[int, int] | None = None) -> None:
         """This modes does not require any transforms."""
         if image_size is not None:
             logger.warning("Ignoring image_size argument as each backend has its own transforms.")
+
+    @staticmethod
+    def _export_not_supported_message() -> None:
+        logging.warning("Exporting the model is not supported for VLM-AD model. Skipping...")
+
+    def to_torch(  # type: ignore[override]
+        self,
+        *_,
+        **__,
+    ) -> None:
+        """Skip export to torch."""
+        return self._export_not_supported_message()
+
+    def to_onnx(  # type: ignore[override]
+        self,
+        *_,
+        **__,
+    ) -> None:
+        """Skip export to onnx."""
+        return self._export_not_supported_message()
+
+    def to_openvino(  # type: ignore[override]
+        self,
+        *_,
+        **__,
+    ) -> None:
+        """Skip export to openvino."""
+        return self._export_not_supported_message()

tests/integration/model/test_models.py

@@ -7,6 +7,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 from pathlib import Path
+from unittest.mock import MagicMock
 
 import pytest
 
@@ -209,6 +210,11 @@ def _get_objects(
             )
 
         model = get_model(model_name, **extra_args)
+
+        if model_name == "vlm_ad":
+            model.vlm_backend = MagicMock()
+            model.vlm_backend.predict.return_value = "YES: Because reasons..."
+
         engine = Engine(
             logger=False,
             default_root_dir=project_path,