Upload app.py

app.py

@@ -0,0 +1,55 @@
+import pandas as pd
+import gradio as gr
+
+# Load final dataset
+df = pd.read_csv("amazon_final_model_dataset.csv")
+
+# Keep only useful columns if they exist
+useful_columns = [
+    "product_name",
+    "avg_rating",
+    "avg_sentiment",
+    "avg_price",
+    "monthly_sales",
+    "monthly_revenue",
+    "monthly_profit",
+    "success_probability",
+    "business_recommendation"
+]
+
+existing_columns = [col for col in useful_columns if col in df.columns]
+df = df[existing_columns].copy()
+
+# Remove duplicates on product name for cleaner dropdown
+df = df.drop_duplicates(subset=["product_name"]).sort_values("product_name")
+
+def analyze_product(product_name):
+    row = df[df["product_name"] == product_name].iloc[0]
+
+    result = f"""
+### Product analysis
+
+**Product:** {row['product_name']}
+
+**Average rating:** {row.get('avg_rating', 'N/A'):.2f}  
+**Average sentiment:** {row.get('avg_sentiment', 'N/A'):.3f}  
+**Average price:** ${row.get('avg_price', 'N/A'):.2f}  
+**Monthly sales:** {int(row.get('monthly_sales', 0))}  
+**Monthly revenue:** ${row.get('monthly_revenue', 0):.2f}  
+**Monthly profit:** ${row.get('monthly_profit', 0):.2f}  
+**Success probability:** {row.get('success_probability', 0):.2%}  
+
+**Business recommendation:**  
+{row.get('business_recommendation', 'No recommendation available')}
+"""
+    return result
+
+demo = gr.Interface(
+    fn=analyze_product,
+    inputs=gr.Dropdown(choices=df["product_name"].tolist(), label="Choose a product"),
+    outputs=gr.Markdown(label="Analysis"),
+    title="Amazon Product Success Analyzer",
+    description="This app analyzes Amazon products using sentiment, ratings, synthetic business metrics, and predicted success probability."
+)
+
+demo.launch()