Diagrams
This section contains visual representations of the genetic algorithm implementation and model architecture within the Ensemble Genetic Algorithm project.
Data Pipeline and Genetic Algorithm
main.py Command-Line Workflow
!main.py Workflow
Source: assets/main_py_workflow.mmd
Description: Illustrates the end-to-end workflow when running an experiment from the command line using
main.py, including the main grid search loop and optional evaluation and plotting steps.
GA Example Usage, Data Grid and GA Grid Permutations, System Flow (example_usage.ipynb)
!GA System Flow
Source: assets/example_usage_permutations.mmd
Description: Illustrates the genetic algorithm search over grid parameters, as demonstrated in the example usage notebook.
GA Data Flow
!GA Data Flow
Source: assets/ga_data_diagram.mmd
Description: Illustrates the flow of data through the genetic algorithm pipeline, from input to ensemble generation.
Model Class Structure
!Model Class Structure
Source: assets/model_classes.mmd
Description: Shows the inheritance hierarchy and relationships between the different model classes used in the project.
Genetic Algorithm Components
Weighting System
!GA Weighting System
Source: assets/ga_weighting.mmd
Description: Demonstrates the weighting mechanism applied to individual base learners within an ensemble.
Parameter Space Grid
!Grid Parameter Space
Source: assets/grid_param_space_ga.mmd
Description: Visualizes how the genetic algorithm explores the parameter space, including feature and hyperparameter grids.
Model Generation Workflows
SVC Model Generation
!SVC Model Generation
Source: assets/svc_model_gen.mmd
Description: Flow diagram detailing the process for generating Support Vector Classifier (SVC) models as base learners.
PyTorch Model Generation
!PyTorch Model Generation
Source: assets/torch_model_gen.mmd
Description: Flow diagram illustrating the generation process for PyTorch neural network models, including aspects of neural architecture search.
XGBoost Model Generation
!XGBoost Model Generation
Source: assets/xgb_model_gen.mmd
Description: Flow diagram outlining the generation process for XGBoost models.
Diagram Format
All diagrams are available in both Mermaid source format (.mmd) and rendered formats (.png/.svg). The Mermaid source files can be edited and re-rendered as needed for documentation updates, ensuring the diagrams remain current with the project’s development.