{"id":28476087,"url":"https://github.com/nmsby/pca-machine-learning-lab","last_synced_at":"2026-05-01T08:31:49.532Z","repository":{"id":297180438,"uuid":"994667408","full_name":"NMsby/pca-machine-learning-lab","owner":"NMsby","description":"Principal Component Analysis (PCA) implementation and analysis lab for Machine Learning. Features manual PCA implementation, scikit-learn applications, data compression, and feature extraction with detailed visualizations.","archived":false,"fork":false,"pushed_at":"2025-06-04T07:08:53.000Z","size":6576,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-06-04T13:57:48.146Z","etag":null,"topics":["data-analysis","dimensionality-reduction","jupyter-notebook","machine-learning","numpy","pca","python","scikit-learn","visualization"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/NMsby.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-06-02T09:46:17.000Z","updated_at":"2025-06-04T09:36:52.000Z","dependencies_parsed_at":"2025-06-04T14:08:04.447Z","dependency_job_id":null,"html_url":"https://github.com/NMsby/pca-machine-learning-lab","commit_stats":null,"previous_names":["nmsby/pca-machine-learning-lab"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/NMsby/pca-machine-learning-lab","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NMsby%2Fpca-machine-learning-lab","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NMsby%2Fpca-machine-learning-lab/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NMsby%2Fpca-machine-learning-lab/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NMsby%2Fpca-machine-learning-lab/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NMsby","download_url":"https://codeload.github.com/NMsby/pca-machine-learning-lab/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NMsby%2Fpca-machine-learning-lab/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":263050679,"owners_count":23405892,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["data-analysis","dimensionality-reduction","jupyter-notebook","machine-learning","numpy","pca","python","scikit-learn","visualization"],"created_at":"2025-06-07T15:06:02.576Z","updated_at":"2026-05-01T08:31:49.520Z","avatar_url":"https://github.com/NMsby.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Principal Component Analysis (PCA) Lab\n\nA comprehensive implementation and analysis of Principal Component Analysis for Machine Learning.\nThis project demonstrates PCA from mathematical foundations to real-world applications.\n\n## 🎯 Project Overview\n\nThis repository contains a complete exploration of PCA including:\n- **Mathematical foundations** and theoretical derivations\n- **From-scratch implementation** using NumPy\n- **Scikit-learn applications** on real datasets\n- **Data compression** and feature extraction examples\n- **Kernel PCA** for nonlinear dimensionality reduction\n- **Comprehensive evaluation** and performance analysis\n\n## 🚀 Features\n\n- ✅ Manual PCA implementation with comprehensive testing\n- ✅ Interactive Jupyter notebooks with detailed explanations\n- ✅ Real-world dataset analysis (Iris, MNIST, Faces)\n- ✅ Data compression with quality analysis\n- ✅ Classification performance comparison\n- ✅ Visualizations and reporting\n\n## 📁 Project Structure\n\n```\npca-machine-learning-lab/\n├── notebooks/              # Interactive analysis notebooks\n│   ├── 01_mathematical_foundations.ipynb\n│   ├── 02_pca_from_scratch.ipynb\n│   ├── 03_scikit_learn_implementation.ipynb\n│   ├── 04_applications.ipynb\n│   └── 05_bonus_kernel_pca.ipynb\n├── src/                    # Source code and utilities\n│   ├── pca_implementation.py\n│   ├── kernel_pca.py\n│   ├── data_utils.py\n│   └── visualization_utils.py\n├── data/                   # Data and results\n│   ├── processed/          # Processed datasets\n│   └── results/            # Analysis results\n├── reports/                # Final report and figures\n│   ├── final_report.pdf\n│   └── figures/\n├── tests/                  # Unit tests\n└── docs/                   # Documentation\n```\n\n## 🔍 Key Results\n\n### Performance Improvements\n- **High-dimensional data (\u003e500D)**: 5-10x speed improvement\n- **Medium-dimensional data (50-500D)**: 2-5x speed improvement\n- **Memory reduction**: 10-50x decrease in memory usage\n- **Accuracy**: Often maintained or improved\n\n### Compression Achievements\n- **Optimal ratios**: 5-50x compression depending on quality requirements\n- **Quality preservation**: \u003e95% correlation with proper component selection\n- **Processing speed**: 200+ images/second on standard hardware\n\n### Kernel PCA Insights\n- **Nonlinear patterns**: 2-5x better class separation\n- **RBF kernel**: Most versatile for unknown patterns\n- **Parameter tuning**: Critical for performance (gamma optimization)\n\n## 🛠️ Installation \u0026 Usage\n\n### Quick Start\n```bash\n# Clone repository\ngit clone https://github.com/NMsby/pca-machine-learning-lab.git\ncd pca-machine-learning-lab\n\n# Create environment\npython -m venv venv\nvenv\\Scripts\\activate  # Windows\n\n# Install dependencies\npip install -r requirements.txt\n\n# Launch Jupyter\njupyter notebook\n```\n\n### Usage Examples\n\n#### Basic PCA Implementation\n```python\nfrom src.pca_implementation import PCA\nimport numpy as np\n\n# Generate sample data\nX = np.random.randn(100, 10)\n\n# Apply PCA\npca = PCA(n_components=3)\nX_transformed = pca.fit_transform(X)\n\nprint(f\"Explained variance ratio: {pca.explained_variance_ratio_}\")\n```\n\n#### Kernel PCA for Nonlinear Data\n```python\nfrom src.kernel_pca import KernelPCA\nfrom sklearn.datasets import make_moons\n\n# Generate nonlinear data\nX, y = make_moons(n_samples=200, noise=0.1)\n\n# Apply Kernel PCA\nkpca = KernelPCA(n_components=2, kernel='rbf', gamma=1.0)\nX_kpca = kpca.fit_transform(X)\n```\n\n## 📊 Datasets Used\n\n- **Iris Dataset** - Classic 4D botanical measurements\n- **MNIST** - Handwritten digit recognition\n- **Olivetti Faces** - Facial recognition dataset\n- **Synthetic Data** - Custom generated for testing\n\n## 📈 Results Summary\n\n### Dataset Analysis\n| Dataset        | Dimensions | Optimal Components | Improvement |\n|----------------|------------|--------------------|-------------|\n| Iris           | 4          | 2 (95.8% variance) | 1.2x speed  |\n| MNIST Digits   | 64         | 15 (90% variance)  | 3.5x speed  |\n| Olivetti Faces | 4,096      | 50 (85% variance)  | 8.2x speed  |\n\n### Application Guidelines\n| Use Case  | Components         | Compression | Priority    |\n|-----------|--------------------|-------------|-------------|\n| Real-time | 5-15% of original  | 5-15x       | Speed       |\n| Storage   | 15-30% of original | 2-8x        | Compression |\n| Analysis  | 30-50% of original | 1-4x        | Quality     |\n\n## 🤝 Contributing\n\nThis is an academic project, but suggestions and improvements are welcome! Please feel free to:\n- Report issues or bugs\n- Suggest improvements to documentation\n- Share interesting use cases or datasets\n- Propose additional features or analyses\n\n## 📄 License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## 🙏 Acknowledgments\n\n- Course materials and lab instructions\n- Scikit-learn documentation and examples\n- Academic papers on PCA methodology\n- Open source community tools and datasets\n\n---\n\n**Author**: Nelson Masbayi  \n**Email**: [nmsby.dev@gmail.com](mailto:nmsby.dev@gmail.com)  \n**Module**: Machine Learning  \n**Institution**: [Strathmore University](https://strathmore.edu)  \n**Date**: June 2025","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnmsby%2Fpca-machine-learning-lab","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnmsby%2Fpca-machine-learning-lab","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnmsby%2Fpca-machine-learning-lab/lists"}