Export Results - Hazelbean Tutorial

Export Results

Learning Time: 10 minutes
Prerequisites: Spatial Analysis
Source File: examples/step_5_export_results.py

Overview

Save processed results and organize outputs in project structure

🎯 Learning Objectives

This tutorial will help you understand:

  • Results Export: Core concepts and practical application

  • Project Organization: Core concepts and practical application

  • Output Management: Core concepts and practical application

πŸ’» Hands-on Learning: Each code block is executable - run them step-by-step to build understanding progressively.

Key Concepts

  • Results Export: Important concept for effective Hazelbean workflows

  • Project Organization: Important concept for effective Hazelbean workflows

  • Output Management: Important concept for effective Hazelbean workflows

Tutorial

πŸš€ Getting Started
  1. Make sure you have the Hazelbean environment activated
  2. Run each code block in order to see the functionality in action
  3. Try modifying parameters to experiment with different results
  4. Refer back to the Key Concepts if you need clarification

πŸ“ Working Directory: This tutorial assumes you’re running from the project root directory.

Complete Example

The following code demonstrates the complete workflow for this step:

"""
Demonstrates proper result export and project organization.
"""

# Initialize project (builds on previous steps)
p = hb.ProjectFlow('hazelbean_tutorial')

print("=== Hazelbean Results Export Demo ===")
print()

# Create some example results to export
print("=== Generating Sample Results ===")

# Create a sample analysis result
rows, cols = 50, 50
result_array = np.random.rand(rows, cols) * 100

# Create a classification result
classification = np.random.choice([1, 2, 3], size=(rows, cols), 
                                p=[0.4, 0.4, 0.2])

print(f"Generated results array: {result_array.shape}")
print(f"Generated classification: {classification.shape}")

# Set up proper output organization
print()
print("=== Organizing Output Structure ===")

# Create timestamped analysis folder
timestamp = datetime.now().strftime("%Y%m%d_%H%M")
analysis_dir = os.path.join(p.output_dir, f'tutorial_analysis_{timestamp}')

# Create subdirectories for different output types
rasters_dir = os.path.join(analysis_dir, 'rasters')
reports_dir = os.path.join(analysis_dir, 'reports')

for directory in [analysis_dir, rasters_dir, reports_dir]:
    os.makedirs(directory, exist_ok=True)
    print(f"βœ“ Created: {os.path.relpath(directory, p.project_dir)}")

# Export raster results (simulate proper geospatial export)
print()
print("=== Exporting Raster Results ===")

# In a real workflow, you'd save with proper geospatial metadata
# For this tutorial, we'll save as simple arrays with documentation

results_file = os.path.join(rasters_dir, 'analysis_results.npy')
np.save(results_file, result_array)
print(f"βœ“ Saved results array: {os.path.basename(results_file)}")

classification_file = os.path.join(rasters_dir, 'classification.npy')
np.save(classification_file, classification)
print(f"βœ“ Saved classification: {os.path.basename(classification_file)}")

# Create metadata file
metadata_file = os.path.join(rasters_dir, 'metadata.txt')
with open(metadata_file, 'w') as f:
    f.write("Raster Metadata\n")
    f.write("===============\n")
    f.write(f"Analysis date: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n")
    f.write(f"Results shape: {result_array.shape}\n")
    f.write(f"Results range: {result_array.min():.2f} to {result_array.max():.2f}\n")
    f.write(f"Classification classes: {sorted(np.unique(classification))}\n")
    f.write("\nClass definitions:\n")
    f.write("1 = Low values\n")
    f.write("2 = Medium values\n") 
    f.write("3 = High values\n")

print(f"βœ“ Created metadata: {os.path.basename(metadata_file)}")

# Generate comprehensive analysis report
print()
print("=== Creating Analysis Report ===")

report_file = os.path.join(reports_dir, 'tutorial_analysis_report.md')

with open(report_file, 'w') as f:
    f.write("# Hazelbean Tutorial Analysis Report\n\n")

    f.write(f"**Analysis Date:** {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n")
    f.write(f"**Project:** {p.project_name}\n\n")

    f.write("## Summary\n\n")
    f.write("This analysis demonstrates the complete Hazelbean workflow from ")
    f.write("project setup through result export.\n\n")

    f.write("## Data Processing Steps\n\n")
    f.write("1. **Project Setup**: Initialized ProjectFlow with organized directories\n")
    f.write("2. **Data Loading**: Used get_path() for intelligent file discovery\n")
    f.write("3. **Processing**: Applied transformations and mathematical operations\n")
    f.write("4. **Analysis**: Performed spatial analysis and multi-raster operations\n")
    f.write("5. **Export**: Organized and documented results\n\n")

    f.write("## Results Summary\n\n")
    f.write(f"- **Results Array**: {result_array.shape} pixels\n")
    f.write(f"- **Value Range**: {result_array.min():.2f} to {result_array.max():.2f}\n")
    f.write(f"- **Mean Value**: {result_array.mean():.2f}\n")
    f.write(f"- **Standard Deviation**: {result_array.std():.2f}\n\n")

    f.write("## Classification Results\n\n")
    for class_val in sorted(np.unique(classification)):
        count = np.sum(classification == class_val)
        percent = (count / classification.size) * 100
        f.write(f"- **Class {class_val}**: {count} pixels ({percent:.1f}%)\n")

    f.write("\n## Files Generated\n\n")
    f.write("- `rasters/analysis_results.npy` - Main analysis results\n")
    f.write("- `rasters/classification.npy` - Classification map\n")
    f.write("- `rasters/metadata.txt` - Raster metadata\n")
    f.write("- `reports/tutorial_analysis_report.md` - This report\n\n")

    f.write("## Project Structure\n\n")
    f.write("```\n")
    f.write(f"{p.project_name}/\n")
    f.write("β”œβ”€β”€ inputs/           # Input data\n")
    f.write("β”œβ”€β”€ intermediate/     # Processing files\n") 
    f.write("└── outputs/          # Final results\n")
    f.write(f"    └── tutorial_analysis_{timestamp}/\n")
    f.write("        β”œβ”€β”€ rasters/  # Geospatial outputs\n")
    f.write("        └── reports/  # Documentation\n")
    f.write("```\n\n")

    f.write("## Next Steps\n\n")
    f.write("- Modify parameters for your own analysis\n")
    f.write("- Use real geospatial data with coordinate systems\n")
    f.write("- Integrate with other Hazelbean functions\n")
    f.write("- Add visualization and plotting\n")

print(f"βœ“ Generated report: {os.path.basename(report_file)}")

# Create quick summary for console
print()
print("=== Project Summary ===")
print(f"Project name: {p.project_name}")
print(f"Project directory: {p.project_dir}")
print(f"Analysis outputs: {os.path.relpath(analysis_dir, p.project_dir)}")

# Count files in each directory
for subdir, name in [(rasters_dir, 'Rasters'), (reports_dir, 'Reports')]:
    file_count = len([f for f in os.listdir(subdir) if os.path.isfile(os.path.join(subdir, f))])
    print(f"{name}: {file_count} files")

print()
print("πŸŽ‰ Tutorial complete! You've learned:")
print("  βœ“ ProjectFlow setup and directory management")
print("  βœ“ Intelligent file discovery with get_path()")
print("  βœ“ Basic raster processing and transformations")
print("  βœ“ Spatial analysis and multi-raster operations") 
print("  βœ“ Professional result organization and documentation")
print()
print(f"πŸ“ Check your results in: {analysis_dir}")
print("πŸ“– Read the full report for next steps and customization ideas")

Step-by-Step Breakdown

Expected Output

When you run this tutorial, you should see output similar to:

Tutorial execution completed successfully.

Learning Tips

πŸ“š Try This Next
  • Modify the example parameters to see how results change
  • Combine concepts from this step with previous tutorials
  • Experiment with your own data using the same patterns

πŸ” Troubleshooting: If you encounter issues, check that your environment is properly activated and all required data files are accessible.