kyriakospsa/classifiers-from-scratch

Implementation using ONLY the Numpy library of classifiers and the evaluation of their performance on the iris plant and the pima Indians diabetes datasets.

Implementation using ONLY the Numpy library of classifiers and the evaluation of their performance on the iris plant and the pima Indians diabetes datasets.

The classifiers implemented are the following:

• A k-nearest neighbours (k-NN) classifier, allowing for the selection of the distance metric between euclidean distance, manhattan distance, Mahalanobis distance and Chebyshev distance.
• A Bayesian classifier, able to utilize different assumptions about the covariance of the features:
  • diagonal covariance matrices
  • non-diagonal covariance matrices
  • Components of the feature vectors that are mutually statistically independent (Naïve Bayes approach)
• Finally, the single-layer perceptron algorithm was implemented.

classifiers-from-scratch's Issues

PyTorch implementation

Implement everything using PyTorch, preferably, in proper modules.

Better README

Improve presentation, highlighting the work done. Maybe add output results and even running instructions. Possible addition of output examples.