require

.github/workflows/ci-cd.yml

@@ -19,6 +19,8 @@ jobs:
       run: |
         pip install --upgrade pip
         pip install -r requirements.txt
+    - name: Install dependencies
+      run: pip install transformers[torch] accelerate>=0.26.0
 
     - name: Train Model
       run: python scripts/train.py

app.py

@@ -1,12 +1,50 @@
 import streamlit as st
-from transformers import pipeline
+import torch
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+import torch.nn.functional as F
 
-# Load trained model
-classifier = pipeline("text-classification", model="./models")
+# Load trained model & tokenizer
+@st.cache_resource
+def load_model():
+    model = AutoModelForSequenceClassification.from_pretrained("models/sentiment_model")
+    tokenizer = AutoTokenizer.from_pretrained("models/sentiment_model")
+    return model, tokenizer
 
-st.title("Sentiment Analysis with DistilBERT")
-text = st.text_area("Enter a text")
+model, tokenizer = load_model()
+
+# Streamlit UI
+st.set_page_config(page_title="Sentiment Analyzer", page_icon="💬", layout="wide")
+
+st.title("💬 Sentiment Analyzer")
+st.write("Analyze the sentiment of any text! Enter a sentence below and get an instant analysis.")
+
+user_input = st.text_area("Enter your text:", "")
+
+if st.button("Analyze Sentiment"):
+    if user_input:
+        with st.spinner("Analyzing..."):
+            inputs = tokenizer(user_input, return_tensors="pt", truncation=True, padding=True)
+            outputs = model(**inputs)
+            probs = F.softmax(outputs.logits, dim=-1)
+            sentiment_index = torch.argmax(probs).item()
+            confidence = round(probs[0][sentiment_index].item() * 100, 2)
+
+        # Map index to label
+        labels = ["Negative", "Neutral", "Positive"]  # Adjust this based on your training labels
+        sentiment = labels[sentiment_index]
+
+        # Display result
+        st.subheader("🔍 Result")
+        if sentiment == "Positive":
+            st.success(f"😊 **Positive Sentiment** ({confidence}%)")
+        elif sentiment == "Negative":
+            st.error(f"😠 **Negative Sentiment** ({confidence}%)")
+        else:
+            st.warning(f"😐 **Neutral Sentiment** ({confidence}%)")
+
+    else:
+        st.warning("⚠️ Please enter some text.")
+
+st.markdown("---")
+st.markdown("🔗 Built with Streamlit | Model: DistilBERT (Fine-tuned)")
 
-if st.button("Predict"):
-    result = classifier(text)
-    st.write(result)