Spaces:

TruthLens
/

fake-real

Runtime error

App Files Files Community

TruthLens commited on Feb 22

Commit

9bdc558

verified ·

1 Parent(s): fdc7914

Update app.py

Browse files

Files changed (1) hide show

app.py +16 -14

app.py CHANGED Viewed

@@ -14,18 +14,14 @@ os.makedirs(upload_folder, exist_ok=True)
 # Updated Fake News Detection Models
 news_models = {
     "mrm8488": pipeline("text-classification", model="mrm8488/bert-tiny-finetuned-fake-news-detection"),
-    "distilbert": pipeline("text-classification", model="distilbert-base-uncased-finetuned-sst-2-english"),
-    "cardiffnlp": pipeline("text-classification", model="cardiffnlp/twitter-roberta-base-offensive")
 }
-# Updated Image Models for AI vs. Human Detection
 clip_model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
 clip_processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
-ai_image_models = {
-    "clip-vit-base-patch32": clip_model
-}
 # Image transformation pipeline
 transform = transforms.Compose([
     transforms.Resize((224, 224)),
@@ -33,7 +29,7 @@ transform = transforms.Compose([
     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
 ])
-# HTML Template with Model Selection
 HTML_TEMPLATE = """
 <!DOCTYPE html>
 <html lang="en">
@@ -42,7 +38,7 @@ HTML_TEMPLATE = """
     <title>AI & News Detection</title>
     <style>
         body { font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; background-color: #f5f5f5; padding: 20px; }
-        .container { background: white; padding: 30px; border-radius: 12px; max-width: 800px; margin: auto; box-shadow: 0 4px 12px rgba(0, 0, 0, 0.1); }
         textarea, select { width: 100%; padding: 12px; margin-top: 10px; border-radius: 8px; border: 1px solid #ccc; }
         button { background-color: #4CAF50; color: white; border: none; padding: 12px 20px; border-radius: 8px; cursor: pointer; font-size: 16px; margin-top: 10px; }
         button:hover { background-color: #45a049; }
@@ -57,8 +53,8 @@ HTML_TEMPLATE = """
             <label for="model">Select Fake News Model:</label>
             <select name="model" required>
                 <option value="mrm8488">MRM8488 (BERT-Tiny)</option>
-                <option value="distilbert">DistilBERT (SST-2)</option>
-                <option value="cardiffnlp">CardiffNLP (RoBERTa-Offensive)</option>
             </select>
             <button type="submit">Detect News Authenticity</button>
         </form>
@@ -80,6 +76,7 @@ HTML_TEMPLATE = """
         <div class="result">
             <h2>📷 Image Detection Result:</h2>
             <p>{{ image_prediction }}</p>
         </div>
         {% endif %}
     </div>
@@ -118,13 +115,18 @@ def detect_image():
     inputs = clip_processor(images=img, return_tensors="pt")
     with torch.no_grad():
-        image_features = ai_image_models["clip-vit-base-patch32"].get_image_features(**inputs)
-    prediction = "AI-Generated" if torch.mean(image_features).item() > 0 else "Human-Created"
     return render_template_string(
         HTML_TEMPLATE,
-        image_prediction=f"Prediction: {prediction}"
     )
 if __name__ == "__main__":

 # Updated Fake News Detection Models
 news_models = {
     "mrm8488": pipeline("text-classification", model="mrm8488/bert-tiny-finetuned-fake-news-detection"),
+    "google-electra": pipeline("text-classification", model="google/electra-base-discriminator"),
+    "bert-base": pipeline("text-classification", model="bert-base-uncased")
 }
+# Updated Image Model for AI vs. Human Detection
 clip_model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
 clip_processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
 # Image transformation pipeline
 transform = transforms.Compose([
     transforms.Resize((224, 224)),
     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
 ])
+# HTML Template with Model Selection and Explanations
 HTML_TEMPLATE = """
 <!DOCTYPE html>
 <html lang="en">
     <title>AI & News Detection</title>
     <style>
         body { font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; background-color: #f5f5f5; padding: 20px; }
+        .container { background: white; padding: 30px; border-radius: 12px; max-width: 850px; margin: auto; box-shadow: 0 4px 12px rgba(0, 0, 0, 0.1); }
         textarea, select { width: 100%; padding: 12px; margin-top: 10px; border-radius: 8px; border: 1px solid #ccc; }
         button { background-color: #4CAF50; color: white; border: none; padding: 12px 20px; border-radius: 8px; cursor: pointer; font-size: 16px; margin-top: 10px; }
         button:hover { background-color: #45a049; }
             <label for="model">Select Fake News Model:</label>
             <select name="model" required>
                 <option value="mrm8488">MRM8488 (BERT-Tiny)</option>
+                <option value="google-electra">Google Electra (Base Discriminator)</option>
+                <option value="bert-base">BERT-Base Uncased</option>
             </select>
             <button type="submit">Detect News Authenticity</button>
         </form>
         <div class="result">
             <h2>📷 Image Detection Result:</h2>
             <p>{{ image_prediction }}</p>
+            <p><strong>Explanation:</strong> The model compares visual features against known AI-generated and human-created image patterns. High feature uniformity often indicates AI generation, while more varied features suggest human-created content.</p>
         </div>
         {% endif %}
     </div>
     inputs = clip_processor(images=img, return_tensors="pt")
     with torch.no_grad():
+        image_features = clip_model.get_image_features(**inputs)
+        mean_feature_value = torch.mean(image_features).item()
+        prediction = "AI-Generated" if mean_feature_value > 0 else "Human-Created"
+    explanation = (
+        f"Prediction: {prediction} (Feature Value: {mean_feature_value:.4f}). "
+        "Higher feature uniformity suggests AI generation, while more variability indicates human creation."
+    )
     return render_template_string(
         HTML_TEMPLATE,
+        image_prediction=explanation
     )
 if __name__ == "__main__":