Update app.py

app.py

@@ -1,28 +1,30 @@
 import streamlit as st
-from transformers import pipeline
-from PIL import Image
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+import torch.nn.functional as F
 
-# Load sentiment analysis pipeline
+# Load model and tokenizer
 @st.cache_resource
 def load_model():
-    return pipeline("sentiment-analysis")
+    model_name = "distilbert-base-uncased-finetuned-sst-2-english"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForSequenceClassification.from_pretrained(model_name)
+    return tokenizer, model
 
-sentiment_analyzer = load_model()
+tokenizer, model = load_model()
 
 # Streamlit UI setup
 st.set_page_config(
     page_title="Mental Health Sentiment Insight",
     page_icon="🧠",
-    layout="centered",
-    initial_sidebar_state="expanded",
+    layout="centered"
 )
 
-# --- Custom CSS for a modern look ---
+# Custom UI
 st.markdown("""
     <style>
         .main {
-            background-color: #f0f2f6;
-            padding: 20px;
+            background-color: #f4f6f9;
         }
         .stTextInput>div>div>input {
             font-size: 16px;
@@ -42,23 +44,26 @@ st.markdown("""
     </style>
 """, unsafe_allow_html=True)
 
-# --- App Header ---
-st.markdown("## 🧠 Mental Health Sentiment Insight")
-st.markdown("Analyze your thoughts to understand your emotional state. This app uses a powerful AI model to detect **Positive**, **Negative**, or **Neutral** sentiment.")
+st.title("🧠 Mental Health Sentiment Insight")
+st.markdown("This app helps analyze your thoughts to detect **Positive** or **Negative** emotions using an AI model.")
 
-# --- User Input ---
-text_input = st.text_area("✍️ Write your thoughts here:", height=200, placeholder="Type something you're feeling...")
+text = st.text_area("✍️ Share your thoughts here:", height=180)
 
 if st.button("Analyze Sentiment"):
-    if text_input.strip() == "":
+    if not text.strip():
         st.warning("Please enter some text to analyze.")
     else:
-        with st.spinner("Analyzing..."):
-            result = sentiment_analyzer(text_input)[0]
-            label = result['label']
-            score = result['score']
+        with st.spinner("Analyzing sentiment..."):
+            # Tokenize and predict
+            inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
+            outputs = model(**inputs)
+            probs = F.softmax(outputs.logits, dim=1)
+            confidence, predicted = torch.max(probs, dim=1)
 
-            # --- Custom Color Feedback ---
-            color = "🟢" if label == "POSITIVE" else "🔴" if label == "NEGATIVE" else "🟡"
+            label = model.config.id2label[predicted.item()]
+            score = confidence.item()
 
-            st.success(f"{color} **Sentiment:** {label} \n\n **Confidence:** {score:.2%}")
+            # Emoji based on label
+            emoji = "🟢" if label == "POSITIVE" else "🔴"
+
+            st.success(f"{emoji} **Sentiment**: {label}\n\n**Confidence**: {score:.2%}")