CNN Hyperparameters Optimization using Metaheuristic Optimizers

This project aims to find the best hyperparameters that can be used in the learning and classification phase that lead to the highest performance metrics using a metaheuristic optimizer. The project uses a public dataset from Kaggle named "COVID-19 Chest X-ray Image Dataset" that can be found at https://www.kaggle.com/dataset....s/alifrahman/covid19 The project utilizes the "Manta-ray Foraging Optimization" (MRFO) metaheuristic optimizer that can be found at https://www.sciencedirect.com/....science/article/pii/

Major Phases
==========================
- Dataset Acquisition.
- Dataset Balancing using Data Augmentation.
- Dataset Scaling (Normalization).
- Dataset Splitting.
- Learning and Optimization.
-- Population Initialization.
-- Fitness Function Evaluation.
-- Population Updating.
-- Logging and Reporting.
-- Repeat the Last Three Steps until Convergence or Iteration Completion.

The fitness function evaluation utilizes the transfer learning CNN model. It runs the learning, testing, and validation of the model using the specified dataset. It reports the required performance metrics and returns the fitness function score.

Project Repo. on GitHub
==========================
Link: https://github.com/HossamBalah....a/CNN-Hyperparameter