Quick Start Guide

Get VariancePro running in minutes with this streamlined guide.

⚡ Launch Applications

Main Application

# Start the main VariancePro application
python app_new.py

# Alternative methods
python app.py
start_app.bat  # Windows batch launcher

Access: http://localhost:7871

Testing Framework (Optional)

# Start the Enhanced NL-to-SQL Testing Framework
python test_enhanced_nl_to_sql_ui.py

# Alternative direct launch
python -c "from nl_to_sql_testing_ui_enhanced import EnhancedNLToSQLTestingUI; EnhancedNLToSQLTestingUI().launch()"

Access: http://localhost:7862

📊 Your First Analysis

1. Upload Data

1. Open VariancePro in your browser: http://localhost:7871
2. Click "Upload CSV File"
3. Select your financial data file
4. Wait for automatic column detection

2. Review Field Mapping

• The system automatically analyzes your data structure
• Review suggested column mappings in the Field Picker section
• Adjust mappings if needed

3. Start Analyzing

Once data is loaded, the chat interface becomes active. Try these example queries:

Basic Queries

"Show me the top 10 products by sales"
"What are our revenue trends this quarter?"
"Compare budget vs actual performance"

Advanced Queries

"Generate an executive summary of Q3 performance"
"Analyze variance and explain external factors"
"Which regions are underperforming and why?"

🎯 Key Features to Try

AI-Powered Analysis

• Chat naturally with your data using the AI interface
• Get executive summaries with expandable details
• Receive actionable insights and recommendations

Market Intelligence

• Automatic news correlation with your data patterns
• Geographic and industry-specific market context
• Real-time business intelligence integration

Multiple Analysis Types

• Contribution Analysis: Find your 80/20 drivers
• Variance Analysis: Budget vs actual comparisons
• Timescale Analysis: Trend and pattern recognition
• Ranking Analysis: Top/bottom performers

📋 Sample Data

If you don't have data ready, VariancePro includes sample datasets:

# Use sample data files
sample_data/comprehensive_sales_data.csv
sample_data/sales_budget_actuals.csv
sample_data/sample_variance_data.csv

🧪 Test the NL-to-SQL Framework

The testing framework lets you experiment with natural language queries:

1. Launch Testing Interface

python test_enhanced_nl_to_sql_ui.py

2. Upload Test Data

• Use any CSV file or the provided sample data
• Framework automatically detects column structure

3. Try Natural Language Queries

"Show me sales where region is North"
"Find records where actual sales > 15000"
"Get data for Q1 2024"
"Show top 3 regions by sales"

4. Compare Translation Strategies

• Test multiple AI models (if available)
• Compare LLM Enhanced vs Semantic Parsing
• View quality scores and performance metrics

⚙️ Quick Configuration

Change Default Ports

# Main application
python app_new.py --port 7872

# Testing framework
# Edit test_enhanced_nl_to_sql_ui.py line with server_port=7863

Select Different AI Model

# Check available models
ollama list

# In the testing framework UI:
# Use the model dropdown to switch between available models

🔍 Verify Everything Works

Health Checks

# Check Ollama is running
curl http://localhost:11434/api/tags

# Validate framework syntax
python validate_nl_to_sql_syntax.py

# Test basic functionality
python test_framework_basic.py

Expected Results

• ✅ Main app loads at http://localhost:7871
• ✅ Testing framework loads at http://localhost:7862
• ✅ File upload works without errors
• ✅ AI chat responds to queries
• ✅ Analysis results display properly

🚨 Quick Troubleshooting

App Won't Start

# Check Python version
python --version  # Should be 3.8+

# Verify dependencies
pip install -r requirements.txt

# Check ports
netstat -an | findstr :7871  # Windows
lsof -i :7871                # macOS/Linux

Ollama Issues

# Start Ollama service
ollama serve

# Pull required model
ollama pull gemma3:latest

Import Errors

# Reinstall dependencies
pip uninstall -r requirements.txt -y
pip install -r requirements.txt

🎯 Next Steps

Once you have VariancePro running:

1. Explore Analysis Types - Understand different analysis capabilities
2. Read Usage Guide - Comprehensive feature documentation
3. Try Analysis Examples - Real-world scenarios
4. Review Best Practices - Optimize your workflow

💡 Pro Tips

• Start with simple queries before trying complex analysis
• Use sample data to familiarize yourself with features
• Check the AI chat history to build on previous questions
• Export results for further analysis or reporting
• Experiment with different models in the testing framework

---

🎉 Congratulations! You're now ready to transform your financial data into strategic intelligence with VariancePro.