championsh / msu-ml-prac

MSU-CMC-SP Machine Learning practicum

MSU-ML-Prac

This repository is devoted to Machine Learning MSU Practicum.

Mastered themes:

1. Handling tabular data using the Pandas library, visualization using the Matplotlib library, Seaborn, Plotly,
2. Vector computation using the NumPy library,
3. K Nearest Neighbors (KNN) algorithm for solving classification and regression tasks,
4. Linear models
   • overtraining experience,
   • Dealing with overtraining,
   • Regularization Techniques,
   • Regression issue;
5. Preprocessing categorical features:
   • One-Hot Encoding
   • Count Encoding
6. Support Vector Machine (SVM):
   • Plotting of nonlinear decision boundary,
   • Optimal selection of the hyperparameter,
   • Principal Component Analysis (PCA) for dimensionality reduction,
   • The Posterior Probability for SVM,
   • Solving ML tasks with the use of SVM, the task was solved with the use of ensemble learning;

Project Tree

├── .gitignore
├── KNN
│   ├── cross_val.py
│   ├── KNN_2023.ipynb
│   └── scalers.py
├── Linear Models: classification
│   ├── Linear_Models_classification .ipynb
│   └── Task.py
├── Linear Models: regression
│   └── Linear_Models_regression.ipynb
├── numpy-pandas-matplotlib
│   ├── functions.py
│   ├── functions_vectorised.py
│   └── Numpy_pandas_matplotlib.ipynb
├── Python Introduction
│   ├── task15.py
│   ├── task6.py
│   └── task7.py
├── README.md
└── SVM
    ├── SVM.ipynb
    └── svm_solution.py