Update app.py

app.py

@@ -2,115 +2,142 @@ import streamlit as st
 import numpy as np
 import tensorflow as tf
 from tensorflow.keras.preprocessing import image
-import pickle
-import requests
-import io
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
 
 # ---------------------------
 # PAGE CONFIG
 # ---------------------------
-st.set_page_config(page_title="TruthGuard AI", layout="wide")
+st.set_page_config(
+    page_title="TruthGuard AI",
+    layout="wide",
+    page_icon="🛡️"
+)
+
+# ---------------------------
+# CUSTOM CSS (UI ENHANCEMENT)
+# ---------------------------
+st.markdown("""
+    <style>
+    .main {
+        background-color: #0E1117;
+        color: white;
+    }
+    .stButton>button {
+        background-color: #4CAF50;
+        color: white;
+        border-radius: 10px;
+        height: 3em;
+        width: 100%;
+    }
+    </style>
+""", unsafe_allow_html=True)
 
+# ---------------------------
+# TITLE
+# ---------------------------
 st.title("🛡️ TruthGuard AI")
-st.subheader("Multi-Modal Fake News & AI Image Detection")
+st.caption("Multi-Modal Fake News & AI Image Detection System")
 
 # ---------------------------
 # LOAD MODELS
 # ---------------------------
 
-# Fake News Model (load from Hugging Face)
+# TEXT MODEL (DistilBERT)
 @st.cache_resource
 def load_text_model():
-    url = "https://huggingface.co/Maheentouqeer1/truthguard-fake-news-detector/resolve/main/fake_news_model.pkl"
-    vectorizer_url = "https://huggingface.co/Maheentouqeer1/truthguard-fake-news-detector/resolve/main/tfidf_vectorizer.pkl"
-
-    model_file = requests.get(url).content
-    vectorizer_file = requests.get(vectorizer_url).content
-
-    model = pickle.loads(model_file)
-    vectorizer = pickle.loads(vectorizer_file)
-
-    return model, vectorizer
+    model_name = "Maheentouqeer1/truthguard-fake-news-detector"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForSequenceClassification.from_pretrained(model_name)
+    return tokenizer, model
+
+# # IMAGE MODEL
+# @st.cache_resource
+# def load_image_model():
+#     model = tf.keras.models.load_model("image_detector_finetuned.h5")
+#     return model
+
+# tokenizer, text_model = load_text_model()
+# image_model = load_image_model()
+import requests
 
-# Image Model
 @st.cache_resource
 def load_image_model():
-    url = "https://huggingface.co/syeda-Rija20/image-detector/resolve/main/image_detector_finetuned.h5"
-    model_file = requests.get(url).content
-
-    with open("temp_model.h5", "wb") as f:
-        f.write(model_file)
+    
+    url = "https://huggingface.co/syeda-Rija20/image-detector/blob/main/image_detector_finetuned.h5"
+    
+    model_path = "image_model.h5"
 
-    model = tf.keras.models.load_model("temp_model.h5")
-    return model
+    # Download model
+    with open(model_path, "wb") as f:
+        f.write(requests.get(url).content)
 
+    # Load model
+    model = tf.keras.models.load_model(model_path)
 
-text_model, vectorizer = load_text_model()
-image_model = load_image_model()
-
+    return model
 # ---------------------------
-# TEXT CLEANING FUNCTION
+# PREDICT TEXT
 # ---------------------------
-import re
-import string
+def predict_news(text):
+    inputs = tokenizer(
+        text,
+        return_tensors="pt",
+        truncation=True,
+        padding=True,
+        max_length=512
+    )
+
+    outputs = text_model(**inputs)
+    probs = torch.nn.functional.softmax(outputs.logits, dim=1)
 
-def clean_text(text):
-    text = text.lower()
-    text = re.sub(r'http\S+', '', text)
-    text = text.translate(str.maketrans('', '', string.punctuation))
-    text = re.sub(r'\d+', '', text)
-    return text
+    prediction = torch.argmax(probs).item()
+    confidence = torch.max(probs).item() * 100
+
+    return prediction, confidence
 
 # ---------------------------
-# TABS UI
+# TABS
 # ---------------------------
 tab1, tab2 = st.tabs(["📰 Fake News Detection", "🖼️ AI Image Detection"])
 
 # ===========================
-# TAB 1 → FAKE NEWS
+# TAB 1 → TEXT
 # ===========================
 with tab1:
-    st.header("Fake News Detector")
+    st.subheader("📰 Fake News Detector")
 
-    user_input = st.text_area("Enter news text:")
+    user_input = st.text_area("Paste news article here...")
 
-    if st.button("Analyze News"):
+    if st.button("🔍 Analyze News"):
         if user_input.strip() == "":
             st.warning("Please enter some text")
         else:
-            clean = clean_text(user_input)
-            vector = vectorizer.transform([clean])
-
-            prediction = text_model.predict(vector)[0]
-            prob = text_model.predict_proba(vector)[0]
-
-            fake_prob = prob[0] * 100
-            real_prob = prob[1] * 100
+            pred, conf = predict_news(user_input)
 
-            if prediction == 0:
-                st.error(f"⚠️ FAKE NEWS ({fake_prob:.2f}%)")
+            if pred == 0:
+                st.error(f"⚠️ FAKE NEWS ({conf:.2f}%)")
             else:
-                st.success(f"✅ REAL NEWS ({real_prob:.2f}%)")
+                st.success(f"✅ REAL NEWS ({conf:.2f}%)")
 
-            st.write("### Confidence Scores")
-            st.write(f"Fake: {fake_prob:.2f}%")
-            st.write(f"Real: {real_prob:.2f}%")
+            st.progress(int(conf))
 
 # ===========================
-# TAB 2 → IMAGE DETECTOR
+# TAB 2 → IMAGE
 # ===========================
 with tab2:
-    st.header("AI Image Detector")
+    st.subheader("🖼️ AI Image Detector")
 
-    uploaded_file = st.file_uploader("Upload Image", type=["jpg", "png"])
+    uploaded_file = st.file_uploader("Upload Image", type=["jpg", "png", "jpeg"])
 
     if uploaded_file is not None:
         img = image.load_img(uploaded_file, target_size=(224, 224))
+        st.image(img, caption="Uploaded Image", use_container_width=True)
+
         img_array = image.img_to_array(img) / 255.0
         img_array = np.expand_dims(img_array, axis=0)
 
         prediction = image_model.predict(img_array)
-
         confidence = float(prediction[0][0]) * 100
 
         if prediction[0][0] > 0.5:
@@ -118,5 +145,10 @@ with tab2:
         else:
             st.success(f"✅ REAL IMAGE ({100-confidence:.2f}%)")
 
-        st.write("### Confidence Score")
-        st.write(f"{confidence:.2f}%")
+        st.progress(int(confidence))
+
+# ---------------------------
+# FOOTER
+# ---------------------------
+st.markdown("---")
+st.caption("Built with ❤️ using Transformers & Deep Learning")