Update app.py

app.py

@@ -1,21 +1,79 @@
 import gradio as gr
 from transformers import pipeline
+import fitz  # PyMuPDF
+import re
+import pandas as pd
 
 # Load models
 bias_detector = pipeline("text-classification", model="himel7/bias-detector")
 bias_type_classifier = pipeline("text-classification", model="maximuspowers/bias-type-classifier")
 
-def detect_bias_and_type(text):
-    detection_result = bias_detector(text)[0]
+def extract_text_from_pdf(pdf_file):
+    """Extract text from a PDF file using PyMuPDF"""
+    text = ""
+    with fitz.open(pdf_file) as pdf:
+        for page in pdf:
+            text += page.get_text("text")
+    return text
+
+def split_into_sentences(text):
+    """Split text into sentences (basic split by .!? with spaces)"""
+    sentences = re.split(r'(?<=[.!?])\s+', text.strip())
+    return [s for s in sentences if s]
+
+def analyze_sentence(sentence):
+    """Run bias detection and (if biased) bias type classification"""
+    detection_result = bias_detector(sentence)[0]
     label = detection_result['label']
     score = detection_result['score']
 
     if label == "LABEL_1":  # Biased
-        type_result = bias_type_classifier(text)[0]
-        return f"🧐 **Bias Detected!**\n- Bias Probability: {score:.2%}\n- Bias Type: {type_result['label']} ({type_result['score']:.2%})"
+        type_result = bias_type_classifier(sentence)[0]
+        return {
+            "sentence": sentence,
+            "bias": "Biased",
+            "bias_score": round(score, 2),
+            "bias_type": type_result['label'],
+            "bias_type_score": round(type_result['score'], 2)
+        }
     else:
-        return f"✅ **Unbiased** (Confidence: {score:.2%})"
+        return {
+            "sentence": sentence,
+            "bias": "Unbiased",
+            "bias_score": round(score, 2),
+            "bias_type": "-",
+            "bias_type_score": "-"
+        }
+
+def analyze_pdf(pdf_file):
+    """Full pipeline: extract text, split sentences, analyze bias"""
+    text = extract_text_from_pdf(pdf_file)
+    sentences = split_into_sentences(text)
+
+    results = [analyze_sentence(s) for s in sentences]
+
+    # Stats
+    total = len(results)
+    biased = sum(1 for r in results if r["bias"] == "Biased")
+    unbiased = total - biased
+
+    stats_md = f"""
+    ### 📊 Bias Statistics
+    - **Total Sentences:** {total}
+    - **Biased Sentences:** {biased} ({(biased/total)*100:.1f}%)
+    - **Unbiased Sentences:** {unbiased} ({(unbiased/total)*100:.1f}%)
+    """
 
+    # Create a DataFrame for table display
+    df = pd.DataFrame(results)
+    return stats_md, df
+
+
+def analyze_text(text):
+    """Single text input analysis"""
+    return analyze_sentence(text)
+
+# Top HTML badges
 badges_html = """
 <p align="center">
   <a href="https://huggingface.co/himel7/bias-detector">
@@ -34,27 +92,21 @@ badges_html = """
 """
 
 with gr.Blocks() as demo:
-    # Render badges using HTML (not escaped)
     gr.HTML(badges_html)
+    gr.Markdown("## Bias Detector + Bias Type Classifier")
 
-    gr.Markdown("## Bias Analyzer")
-    gr.Markdown("Enter a sentence to detect bias and its type.")
-
-    text_input = gr.Textbox(lines=3, placeholder="Enter a sentence...")
-    output = gr.Markdown()
-
-    btn = gr.Button("Analyze")
-    btn.click(detect_bias_and_type, inputs=text_input, outputs=output)
+    with gr.Tab("Single Sentence"):
+        text_input = gr.Textbox(lines=3, placeholder="Enter a sentence...")
+        output = gr.JSON()
+        btn = gr.Button("Analyze")
+        btn.click(analyze_text, inputs=text_input, outputs=output)
 
-    # Examples
-    gr.Examples(
-        examples=[
-            ["The brilliant leader saved the country from disaster."],
-            ["The government announced new tax reforms."],
-            ["The selfish billionaire hoarded his wealth."]
-        ],
-        inputs=text_input
-    )
+    with gr.Tab("Analyze PDF"):
+        pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
+        stats_output = gr.Markdown()
+        table_output = gr.Dataframe(headers=["Sentence", "Bias", "Bias Score", "Bias Type", "Bias Type Score"])
+        analyze_btn = gr.Button("Analyze PDF")
+        analyze_btn.click(analyze_pdf, inputs=pdf_input, outputs=[stats_output, table_output])
 
 if __name__ == "__main__":
     demo.launch()