Update streamlit_app.py

streamlit_app.py

@@ -43,68 +43,68 @@ st.sidebar.markdown("### About\nThis tool uses pretrained models to detect text
 if module == "Text Misinformation":
     st.title("📰 Misinformation Detection")
 
-    user_input = st.text_area("Enter a news statement or claim:", height=150)
-
-    if st.button("Analyze Text"):
-        if not user_input.strip():
-            st.warning("Please enter some text to analyze.")
-        else:
-            model = load_text_model()
-            with st.spinner("Running model inference..."):
-                result = model(user_input, truncation=True)[0]
-
-            label = result.get("label", "N/A")
-            score = float(result.get("score", 0.0))
-
-            # Map labels to human-readable format
-            label_map = {"LABEL_0": "REAL", "LABEL_1": "FAKE", "REAL": "REAL", "FAKE": "FAKE"}
-            readable_label = label_map.get(label, label)
-
-            st.success(f"Prediction: {readable_label}")
-            st.metric("Confidence", f"{score*100:.2f}%")
-
-            # Show raw model output for debugging
-            with st.expander("🔍 Raw Model Output"):
-                st.json(result)
-
-            st.session_state.setdefault("history", [])
-            if len(st.session_state.history) < 20:
-                st.session_state.history.append({
-                    "type": "text",
-                    "input": user_input,
-                    "label": readable_label,
-                    "score": score,
-                    "time": datetime.utcnow().isoformat()
-                })
+    with st.container():
+        user_input = st.text_area("Enter a news statement or claim:", height=150)
+
+        if st.button("Analyze Text"):
+            if not user_input.strip():
+                st.warning("Please enter some text to analyze.")
+            else:
+                model = load_text_model()
+                with st.spinner("Running model inference..."):
+                    result = model(user_input, truncation=True)[0]
+
+                label = result.get("label", "N/A")
+                score = float(result.get("score", 0.0))
+
+                label_map = {"LABEL_0": "REAL", "LABEL_1": "FAKE", "REAL": "REAL", "FAKE": "FAKE"}
+                readable_label = label_map.get(label, label)
+
+                st.success(f"Prediction: {readable_label}")
+                st.metric("Confidence", f"{score*100:.2f}%")
+
+                with st.expander("🔍 Raw Model Output"):
+                    st.json(result)
+
+                st.session_state.setdefault("history", [])
+                if len(st.session_state.history) < 20:
+                    st.session_state.history.append({
+                        "type": "text",
+                        "input": user_input,
+                        "label": readable_label,
+                        "score": score,
+                        "time": datetime.utcnow().isoformat()
+                    })
 
 # ------------------- Image Deepfake -------------------
 elif module == "Image Deepfake":
     st.title("🖼 Deepfake Image Detection")
 
-    uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
+    with st.container():
+        uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
 
-    if uploaded_file:
-        image = Image.open(uploaded_file).convert("RGB")
-        st.image(image, caption="Uploaded Image", use_container_width=True)
+        if uploaded_file:
+            image = Image.open(uploaded_file).convert("RGB")
+            st.image(image, caption="Uploaded Image", use_container_width=True)
 
-        if st.button("Analyze Image"):
-            model = load_image_model()
-            with st.spinner("Running image model..."):
-                results = model(image, top_k=3)
+            if st.button("Analyze Image"):
+                model = load_image_model()
+                with st.spinner("Running image model..."):
+                    results = model(image, top_k=3)
 
-            df = pd.DataFrame([{"label": r.get("label"), "score": r.get("score")} for r in results])
-            st.dataframe(df)
+                df = pd.DataFrame([{"label": r.get("label"), "score": r.get("score")} for r in results])
+                st.dataframe(df)
 
-            best = results[0]
-            st.session_state.setdefault("history", [])
-            if len(st.session_state.history) < 20:
-                st.session_state.history.append({
-                    "type": "image",
-                    "filename": getattr(uploaded_file, "name", "upload"),
-                    "label": best.get("label"),
-                    "score": float(best.get("score", 0.0)),
-                    "time": datetime.utcnow().isoformat()
-                })
+                best = results[0]
+                st.session_state.setdefault("history", [])
+                if len(st.session_state.history) < 20:
+                    st.session_state.history.append({
+                        "type": "image",
+                        "filename": getattr(uploaded_file, "name", "upload"),
+                        "label": best.get("label"),
+                        "score": float(best.get("score", 0.0)),
+                        "time": datetime.utcnow().isoformat()
+                    })
 
 # ------------------- History -------------------
 if "history" in st.session_state and st.session_state.history:
@@ -113,4 +113,3 @@ if "history" in st.session_state and st.session_state.history:
         st.dataframe(df_hist)
         csv = df_hist.to_csv(index=False).encode("utf-8")
         st.download_button("Download History (CSV)", data=csv, file_name="analysis_history.csv", mime="text/csv")
-