/ml-techniques

Primary LanguageJupyter NotebookMIT LicenseMIT

ml-techniques

A collection of machine learning codes in python's numpy and tensorflow.

Techniques

Clustering

  • k-means [19 Dec 2020]
    • k-means with farthest-first heuristic (ffh) [11 Jan 2021]
    • k-means++ [10 Jan 2021]
  • Spectral Clustering
    • Unnormalized Spectral Clustering [23 Dec 2020]

Manifold Learning

  • PCA [9 Jan 2021]
  • Classical MDS [22 Jan 2021]
  • Fast ICA [23 Jan 2021]

Linear Models

  • Logistic Regression [13 Jan 2021]
    • l2 regularization [15 Jan 2021]
  • Linear Regression [16 Jan 2021]
    • Ridge Regression [18 Jan 2021]
    • LASSO [20 Jan 2021]

Bayesian Models

  • Bayesian Linear Regression [19 Jan 2021]

Sampling Methods

  • Importance Sampling [21 Jan 2021]

Requirements

Install requirements.txt file to make sure correct versions of libraries are being used.

  • Python 3.8.x
  • Jupyterlab==2.2.9
  • Numpy==1.19.4
  • Opencv-python==4.4.0.46
  • Pandas==1.1.1
  • Seaborn==0.11.0
  • Scikit-learn==0.24.0
  • Tensorflow==2.4.0
  • Tensorflow-datasets==4.1.0
  • Tensorflow-probability==0.11.1
  • cvxpy==1.1.7

Resources

  • DaumĂ© III, Hal. "A course in machine learning." Publisher, ciml. info 5 (2012): 69.
  • Bishop, Christopher M. Pattern recognition and machine learning. springer, 2006.
  • Murphy, Kevin P. Machine learning: a probabilistic perspective. MIT press, 2012.
  • Miller, Jeffrey W. mathematicalmonk youtube lecture study page Machine Learning Playlist.

License

The MIT License (MIT)

Copyright (c) 2020 Peratham Wiriyathammabhum