Create app.py

app.py

@@ -0,0 +1,50 @@
+# app.py
+import streamlit as st
+from transformers import pipeline
+
+# Title
+st.title("🎬 Movie Review Sentiment Classifier")
+
+# Load model from Hugging Face Hub
+@st.cache_resource
+def load_model():
+    return pipeline("sentiment-analysis", model="Gamer-Dude-77/my-imdb-sentiment-model")
+
+classifier = load_model()
+
+# Text input
+st.subheader("Enter a Review")
+text_input = st.text_area("Type or paste your movie review below:", height=150)
+
+# Prediction
+if st.button("Analyze Sentiment"):
+    if text_input.strip():
+        results = classifier([text_input])
+        result = results[0]
+
+        st.write("### 📊 Prediction Result")
+        if result["label"].upper() == "POSITIVE":
+            st.success(f"😀 Positive (Confidence: {result['score']:.4f})")
+        elif result["label"].upper() == "NEGATIVE":
+            st.error(f"😞 Negative (Confidence: {result['score']:.4f})")
+        else:
+            st.info(f"😐 Neutral (Confidence: {result['score']:.4f})")
+    else:
+        st.warning("⚠️ Please enter some text to analyze.")
+
+# Batch testing
+st.subheader("Batch Testing")
+uploaded_file = st.file_uploader("Upload a .txt file with one review per line", type=["txt"])
+
+if uploaded_file is not None:
+    lines = uploaded_file.read().decode("utf-8").splitlines()
+    if st.button("Analyze File"):
+        results = classifier(lines)
+        for review, res in zip(lines, results):
+            if res["label"].upper() == "POSITIVE":
+                st.success(f"Review: {review}\n→ 😀 Positive ({res['score']:.4f})")
+            elif res["label"].upper() == "NEGATIVE":
+                st.error(f"Review: {review}\n→ 😞 Negative ({res['score']:.4f})")
+            else:
+                st.info(f"Review: {review}\n→ 😐 Neutral ({res['score']:.4f})")
+            st.markdown("---")