UPD: fixed rgb + decreased size

giskard_vision/detectors/base.py

@@ -111,6 +111,7 @@ def get_issues(
                         group=IssueGroup(result.group, "Warning"),
                         meta=result.get_meta_required(),
                         example_manager=ExamplesImages,
+                        display_warnings=False
                     )
                     current_issue.add_examples(result.filename_examples)
                     issues.append(current_issue)

giskard_vision/landmark_detection/dataloaders/base.py

@@ -503,6 +503,8 @@ class DataLoaderWrapper(DataIteratorBase):
         _wrapped_dataloader (DataIteratorBase): The wrapped data loader instance.
     """
 
+    dataloader_type: str = "standard"
+
     def __init__(self, dataloader: DataIteratorBase) -> None:
         """
         Initializes the DataLoaderWrapper with a given DataIteratorBase instance.

giskard_vision/landmark_detection/detectors/base.py

@@ -6,7 +6,10 @@
 from giskard_vision.landmark_detection.tests.performance import NMEMean
 from giskard_vision.utils.errors import GiskardImportError
 
+import os
+import cv2
 
+
 class LandmarkDetectionBaseDetector(DetectorVisionBase):
     """
     Abstract class for Landmark Detection Detectors
@@ -38,11 +41,29 @@ def get_results(self, model: Any, dataset: Any) -> Sequence[ScanResult]:
                 dataloader_ref=dataset,
             )
 
-            results.append(self.get_scan_result(test_result))
+            # Save example images from dataloader and dataset
+            os.makedirs("examples_images", exist_ok=True)
+            filename_examples = []
+
+            filename_example_dataloader_ref = f"examples_images/{dataset.name}.png"
+            cv2.imwrite(
+                filename_example_dataloader_ref,
+                cv2.resize(dataset[0][0][0], (0, 0), fx=0.3, fy=0.3)
+            )
+            filename_examples.append(filename_example_dataloader_ref)
+
+            filename_example_dataloader = f"examples_images/{dl.name}.png"
+            cv2.imwrite(
+                filename_example_dataloader,
+                cv2.resize(dl[0][0][0], (0, 0), fx=0.3, fy=0.3)
+            )
+            filename_examples.append(filename_example_dataloader)
+
+            results.append(self.get_scan_result(test_result, filename_examples))
 
         return results
 
-    def get_scan_result(self, test_result) -> ScanResult:
+    def get_scan_result(self, test_result, filename_examples) -> ScanResult:
         try:
             from giskard.scanner.issues import IssueLevel
         except (ImportError, ModuleNotFoundError) as e:
@@ -63,5 +84,5 @@ def get_scan_result(self, test_result) -> ScanResult:
             metric_reference_value=test_result.metric_value_ref,
             issue_level=issue_level,
             slice_size=test_result.size_data,
-            filename_examples=test_result.filename_examples,
+            filename_examples=filename_examples,
         )

giskard_vision/landmark_detection/tests/base.py

@@ -52,7 +52,6 @@ class TestResult:
     dataloader_ref_name: Optional[str] = None
     size_data: Optional[int] = None
     issue_name: Optional[str] = None
-    filename_examples: Optional[str] = None
 
     def _repr_html_(self):
         """
@@ -314,21 +313,6 @@ def run(
             [prediction_result.prediction_fail_rate, prediction_result_ref.prediction_fail_rate]
         )
 
-        import os
-
-        import matplotlib.pyplot as plt
-
-        os.makedirs("examples_images", exist_ok=True)
-        filename_examples = []
-
-        filename_example_dataloader_ref = f"examples_images/{dataloader_ref.name}.png"
-        plt.imsave(filename_example_dataloader_ref, dataloader_ref[0][0][0])
-        filename_examples.append(filename_example_dataloader_ref)
-
-        filename_example_dataloader = f"examples_images/{dataloader.name}.png"
-        plt.imsave(filename_example_dataloader, dataloader[0][0][0])
-        filename_examples.append(filename_example_dataloader)
-
         return TestResult(
             test_name=self.__class__.__name__,
             description=self.metric.description,
@@ -347,5 +331,4 @@ def run(
             dataloader_ref_name=dataloader_ref.name,
             size_data=len(dataloader),
             issue_name=dataloader.name,
-            filename_examples=filename_examples,
         )

giskard_vision/scanner/scanner.py

@@ -85,8 +85,6 @@ def analyze(
                 detectors, model, dataset, verbose=verbose, raise_exceptions=raise_exceptions
             )
 
-        # issues = self._postprocess(issues)
-
         # Scan completed
         elapsed = perf_counter() - time_start
 
@@ -126,14 +124,6 @@ def _run_detectors(self, detectors, model, dataset, verbose=True, raise_exceptio
 
         return issues, errors
 
-    def _postprocess(self, issues: Sequence[Issue]) -> Sequence[Issue]:
-        # If we detected a Stochasticity issue, we will have a possibly false
-        # positive DataLeakage issue. We remove it here.
-        if any(issue.group == Stochasticity for issue in issues):
-            issues = [issue for issue in issues if issue.group != DataLeakage]
-
-        return issues
-
     def get_detectors(self, tags: Optional[Sequence[str]] = None) -> Sequence:
         """Returns the detector instances."""
         detectors = []