A Python-based machine learning toolkit that implements and evaluates multiple classification algorithms on different datasets. The repository demonstrates the application of various classifier implementations from scikit-learn, including K-Nearest Neighbors, Multi-layer Perceptron, AdaBoost, Bagging, and Gradient Boosting.
-
Implementation of multiple classification algorithms:
- K-Nearest Neighbors with automatic parameter optimization
- Multi-layer Perceptron Neural Network
- AdaBoost Classifier
- Bagging Classifier
- Gradient Boosting Classifier
-
Built-in evaluation metrics and cross-validation
-
Support for multiple datasets:
- Digits dataset (built-in scikit-learn dataset)
- 20 Newsgroups text classification dataset
-
Text preprocessing using TF-IDF vectorization
-
Model performance evaluation using various metrics
scikit-learn
numpy