Update app.py

app.py

@@ -72,6 +72,59 @@ def explain_sql_output(sql_query, query_result):
     )
     return chat_completion.choices[0].message.content.strip()
 
+### --- TAB 3: Data Summary & Insights --- ###
+
+def summarize_data(data_file):
+    try:
+        df = pd.read_csv(data_file.name)
+    except Exception:
+        try:
+            df = pd.read_json(data_file.name)
+        except Exception as e:
+            return f"Failed to read file: {str(e)}"
+
+    metadata = "\n".join([f"- {col}: {str(dtype)}" for col, dtype in df.dtypes.items()])
+    preview = df.head(30).to_csv(index=False)
+
+    system_prompt = """
+You are a professional data analyst AI that specializes in summarizing datasets and uncovering insights.
+
+Your task is to:
+1. Provide a high-level summary of the dataset.
+2. Identify 5 insightful observations or trends.
+3. Highlight any interesting patterns, anomalies, or correlations you find.
+4. Use simple, clear language understandable to non-technical users.
+5. If relevant, suggest what kind of decisions or actions could be made based on the data.
+
+Only use the provided preview and schema — do not assume missing data or guess columns.
+Be helpful, concise, and specific.
+"""
+
+    user_prompt = f"""
+Here is a preview of the dataset (first 30 rows):
+
+{preview}
+
+Here is the schema (column name: type):
+
+{metadata}
+
+Please generate:
+- A brief overview of the dataset.
+- Five unique, data-driven insights or summaries.
+- Optional: Patterns or anomalies worth noting.
+- Keep it understandable and actionable.
+"""
+
+    chat_completion = client.chat.completions.create(
+        messages=[
+            {"role": "system", "content": system_prompt.strip()},
+            {"role": "user", "content": user_prompt.strip()}
+        ],
+        model="llama3-70b-8192"
+    )
+
+    return chat_completion.choices[0].message.content.strip()
 
 ### --- Gradio Interface --- ###
 tab1 = gr.Interface(
@@ -93,6 +146,14 @@ tab2 = gr.Interface(
     description="Input a SQL query and its result. Get an AI-generated explanation."
 )
 
+tab3 = gr.Interface(
+    fn=summarize_data,
+    inputs=gr.File(label="Upload CSV or JSON Dataset"),
+    outputs="text",
+    title="Data Summary & Insights (Groq + LLaMA3)",
+    description="Upload a dataset to get a general summary and 5 AI-generated insights."
+)
+
 demo = gr.TabbedInterface([tab1, tab2], ["SQL Generator", "Explain Output"])
 
 if __name__ == '__main__':