Update deepfake_vs_real.py

deepfake_vs_real.py

@@ -1,14 +1,12 @@
 import gradio as gr
-from transformers import AutoImageProcessor
-from transformers import SiglipForImageClassification
-from transformers.image_utils import load_image
+from transformers import ViTForImageClassification, ViTImageProcessor
 from PIL import Image
 import torch
 
-# Load model and processor
-model_name = "prithivMLmods/Deepfake-vs-Real-8000"
-model = SiglipForImageClassification.from_pretrained(model_name)
-processor = AutoImageProcessor.from_pretrained(model_name)
+# Load the model and processor
+model_name = "prithivMLmods/Deep-Fake-Detector-v2-Model"
+model = ViTForImageClassification.from_pretrained(model_name)
+processor = ViTImageProcessor.from_pretrained(model_name)
 
 def deepfake_classification(image):
     """Predicts whether an image is a Deepfake or Real."""
@@ -18,20 +16,17 @@ def deepfake_classification(image):
     with torch.no_grad():
         outputs = model(**inputs)
         logits = outputs.logits
-        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
+        predicted_class = torch.argmax(logits, dim=1).item()
     
-    labels = {
-        "0": "Deepfake", "1": "Real one"
-    }
-    predictions = {labels[str(i)]: round(probs[i], 3) for i in range(len(probs))}
-    
-    return predictions
+    # Get label mapping
+    label = model.config.id2label[predicted_class] if hasattr(model.config, "id2label") else str(predicted_class)
+    return {label: 1.0}  # Gradio Label output expects a dictionary
 
 # Create Gradio interface
 iface = gr.Interface(
     fn=deepfake_classification,
     inputs=gr.Image(type="numpy"),
-    outputs=gr.Label(label="Prediction Scores"),
+    outputs=gr.Label(label="Prediction"),
     title="Deepfake vs. Real Image Classification",
     description="Upload an image to determine if it's a Deepfake or a Real one."
 )