Update app.py

app.py

@@ -20,7 +20,7 @@ if not openai_api_key:
     st.stop()
 
 # Step 1: Upload CSV data file (or use default)
-st.title("Natural Language to SQL Query App with Enhanced Insights")
+st.title("Natural Language to SQL Query App with Dynamic Insights")
 st.write("Upload a CSV file to get started, or use the default dataset.")
 
 csv_file = st.file_uploader("Upload your CSV file", type=["csv"])
@@ -71,29 +71,29 @@ sql_generation_chain = LLMChain(llm=llm, prompt=sql_prompt)
 
 # Insights Generation Chain
 insights_template = """
-You are an expert data scientist. Based on the user's question and the SQL query result provided below, generate a concise and informative analysis that includes data insights and actionable recommendations.
+You are an expert data scientist. Based on the user's question and the SQL query result provided below, generate a concise and informative analysis that includes data insights.
 
 User's Question: {question}
 
 SQL Query Result:
 {result}
 
-Analysis and Recommendations:
+Analysis:
 """
 insights_prompt = PromptTemplate(template=insights_template, input_variables=['question', 'result'])
 insights_chain = LLMChain(llm=llm, prompt=insights_prompt)
 
-# General Insights and Recommendations Chain
-general_insights_template = """
-You are an expert data scientist. Based on the entire dataset provided below, generate a comprehensive analysis that includes key insights and actionable recommendations.
+# Recommendations Generation Chain
+recommendations_template = """
+You are an expert data scientist. Based on the dataset summary provided below, generate actionable recommendations for improving performance.
 
 Dataset Summary:
 {dataset_summary}
 
-Analysis and Recommendations:
+Recommendations:
 """
-general_insights_prompt = PromptTemplate(template=general_insights_template, input_variables=['dataset_summary'])
-general_insights_chain = LLMChain(llm=llm, prompt=general_insights_prompt)
+recommendations_prompt = PromptTemplate(template=recommendations_template, input_variables=['dataset_summary'])
+recommendations_chain = LLMChain(llm=llm, prompt=recommendations_prompt)
 
 # Optional: Clean up function to remove incorrect COLLATE NOCASE usage
 def clean_sql_query(query):
@@ -121,7 +121,7 @@ def clean_sql_query(query):
 def classify_query(question):
     """Classify the user query as either 'SQL' or 'INSIGHTS'."""
     classification_template = """
-    You are an AI assistant that classifies user queries into two categories: 'SQL' for specific data retrieval queries and 'INSIGHTS' for general analytical or recommendation queries.
+    You are an AI assistant that classifies user queries into two categories: 'SQL' for specific data retrieval queries and 'INSIGHTS' for general analytical queries.
 
     Determine the appropriate category for the following user question.
 
@@ -178,19 +178,8 @@ def process_input():
                 }).strip()
 
                 if generated_sql.upper() == "NO_SQL":
-                    # Handle cases where no SQL should be generated
-                    assistant_response = "Sure, let's discuss some general insights and recommendations based on the data."
-                    
-                    # Generate dataset summary
-                    dataset_summary = generate_dataset_summary(data)
-                    
-                    # Generate general insights and recommendations
-                    general_insights = general_insights_chain.run({
-                        'dataset_summary': dataset_summary
-                    })
-                    
-                    # Append the assistant's insights to the history
-                    st.session_state.history.append({"role": "assistant", "content": general_insights})
+                    assistant_response = "This query is too vague for generating SQL. Please ask a more specific question."
+                    st.session_state.history.append({"role": "assistant", "content": assistant_response})
                 else:
                     # Clean the SQL query
                     cleaned_sql = clean_sql_query(generated_sql)
@@ -207,7 +196,7 @@ def process_input():
                             # Convert the result to a string for the insights prompt
                             result_str = result.head(10).to_string(index=False)  # Limit to first 10 rows
 
-                            # Generate insights and recommendations based on the query result
+                            # Generate insights based on the query result
                             insights = insights_chain.run({
                                 'question': user_prompt,
                                 'result': result_str
@@ -222,16 +211,29 @@ def process_input():
                         assistant_response = f"Error executing SQL query: {e}"
                         st.session_state.history.append({"role": "assistant", "content": assistant_response})
             else:  # INSIGHTS category
-                # Generate dataset summary
-                dataset_summary = generate_dataset_summary(data)
+                if "recommendations" in user_prompt.lower():
+                    # Generate dataset summary for recommendations
+                    dataset_summary = generate_dataset_summary(data)
 
-                # Generate general insights and recommendations
-                general_insights = general_insights_chain.run({
-                    'dataset_summary': dataset_summary
-                })
+                    # Generate recommendations based on the dataset summary
+                    recommendations = recommendations_chain.run({
+                        'dataset_summary': dataset_summary
+                    })
 
-                # Append the assistant's insights to the history
-                st.session_state.history.append({"role": "assistant", "content": general_insights})
+                    # Append the assistant's recommendations to the history
+                    st.session_state.history.append({"role": "assistant", "content": recommendations})
+                else:
+                    # Generate dataset summary for general insights (without recommendations)
+                    dataset_summary = generate_dataset_summary(data)
+
+                    # Generate general insights
+                    general_insights = insights_chain.run({
+                        'question': user_prompt,
+                        'result': dataset_summary
+                    })
+
+                    # Append the assistant's general insights to the history
+                    st.session_state.history.append({"role": "assistant", "content": general_insights})
         
         except Exception as e:
             logging.error(f"An error occurred: {e}")