Skip to content

Latest commit

 

History

History
12 lines (9 loc) · 667 Bytes

File metadata and controls

12 lines (9 loc) · 667 Bytes

Hand-coded-Machine-Learning-Algorithms

Implement machine learning algorithms while trying my best to avoid the usage of machine learning libaries such as sklearn, only use numpy, scipy

Contents

  • Linear Algebra: Gaussian Elimination and Gram-Schmidt orthogonalization
  • K-means: sklearn was used to plot the Elbow method, which is optional
  • Polynomial Regression Using Gradient Descent
  • GMM: sklearn was used to initialize parameters for simplicity, actually I can use the hand-made K-means, or just use a set of randomly initialized parameters
  • PCA of High Dementional Data

Note: encapsulation needed and will be done in the future