How fionaverzivolli/exploring-ge-pivoting-and-lu-factorization works
Main Components
• Algorithm Implementations:
• GE_no_pivot.py (No Pivoting)
• GE_partial_pivot.py (Partial Pivoting)
• GE_complete_pivoting.py (Complete Pivoting)
• Testing Module:
• GE_testing.py (Unit tests using NumPy for correctness and functionality)
• Visualization Module:
• GE_visualize.py (Generates plots and visual performance metrics using Pandas)
• Documentation:
• README.md (Describes the purpose, underlying methodology, and comparison of the three algorithms)
Representing Relationships and Interactions
• Input Data Flow:
• Algorithm Execution:
• Testing Flow:
• The testing module validates the correctness and robustness of each method.
• Visualization Flow:
• Also, indicate that performance metrics (like residuals, conditioning numbers, error metrics, etc.) from the algorithm executions are gathered and then passed to GE_visualize.py.
• This module then outputs relevant visualizations/charts.
Architecturally Relevant Patterns
• Modularity:
• Data Flow Architecture:
• Dependence on Third-Party Libraries:
• External dependencies such as NumPy (for numerical testing) and Pandas (for visualization) near the testing and visualization modules.