oof...

app.py

@@ -122,7 +122,7 @@ register_model_with_metadata(
 )
 
 # --- ONNX Quantized Model Example ---
-ONNX_QUANTIZED_MODEL_PATH = "./models/model_1_quantized.onnx" # Placeholder for your ONNX model
+ONNX_QUANTIZED_MODEL_PATH = "./models/model_1_quantized.onnx"
 
 def preprocess_onnx_input(image: Image.Image):
     # Preprocess image for ONNX model (e.g., for SwinV2, usually 256x256, normalized)
@@ -423,11 +423,11 @@ def full_prediction(img, confidence_threshold, rotate_degrees, noise_level, shar
     gradient_image2 = gradient_processing(img_np_og, intensity=45, equalize=True)
     minmax_image = minmax_process(img_np_og)
     minmax_image2 = minmax_process(img_np_og, radius=6)
-    bitplane_image = bit_plane_extractor(img_pil)
+    # bitplane_image = bit_plane_extractor(img_pil)
     ela1 = ELA(img_np_og, quality=75, scale=50, contrast=20, linear=False, grayscale=True)
     ela2 = ELA(img_np_og, quality=75, scale=75, contrast=25, linear=False, grayscale=True)
     ela3 = ELA(img_np_og, quality=75, scale=75, contrast=25, linear=False, grayscale=False)
-    forensics_images = [img_pil, ela1, ela2, ela3, gradient_image, gradient_image2, minmax_image, minmax_image2, bitplane_image]
+    forensics_images = [img_pil, ela1, ela2, ela3, gradient_image, gradient_image2, minmax_image, minmax_image2]
     forensic_output_descriptions = [
         f"Original augmented image (PIL): {img_pil.width}x{img_pil.height}",
         "ELA analysis (Pass 1): Grayscale error map, quality 75.",
@@ -437,7 +437,7 @@ def full_prediction(img, confidence_threshold, rotate_degrees, noise_level, shar
         "Gradient processing: Int=45, Equalize=True",
         "MinMax processing: Deviations in local pixel values.",
         "MinMax processing (Radius=6): Deviations in local pixel values.",
-        "Bit Plane extractor: Visualization of individual bit planes from different color channels."
+        # "Bit Plane extractor: Visualization of individual bit planes from different color channels."
     ]
     anomaly_detection_results = anomaly_agent.analyze_forensic_outputs(forensic_output_descriptions)
     logger.info(f"Forensic anomaly detection: {anomaly_detection_results['summary']}")
@@ -554,7 +554,7 @@ def predict(img):
         handle_file(img),
         api_name="/simple_predict"
     )
-    return result
+    return str(result)
 community_forensics_preview = gr.Interface(
     fn=predict,
     inputs=gr.Image(type="filepath"),