Skip to content

A collection of tutorials and examples for solving and understanding machine learning and pattern classification tasks

License

Notifications You must be signed in to change notification settings

gunturus/pattern_classification

 
 

Repository files navigation

logo


**Tutorials, examples, collections, and everything else that falls into the categories: pattern classification, machine learning, and data mining.**



Sections



[Download a PDF version] of this flowchart.






Introduction to Machine Learning and Pattern Classification

[back to top]

  • Predictive modeling, supervised machine learning, and pattern classification - the big picture [Markdown]

  • Entry Point: Data - Using Python's sci-packages to prepare data for Machine Learning tasks and other data analyses [IPython nb]

  • An Introduction to simple linear supervised classification using scikit-learn [IPython nb]






Pre-processing

[back to top]

  • Scaling and Normalization

    • About Feature Scaling: Standardization and Min-Max-Scaling (Normalization) [IPython nb]
  • Feature Selection

    • Sequential Feature Selection Algorithms [IPython nb]
  • Dimensionality Reduction

    • Principal Component Analysis (PCA) [IPython nb]

    • The effect of scaling and mean centering of variables prior to a PCA [PDF] [HTML]

    • PCA based on the covariance vs. correlation matrix [IPython nb]

    • Linear Discriminant Analysis (LDA) [IPython nb]

    • Kernel tricks and nonlinear dimensionality reduction via PCA [IPython nb]




Model Evaluation

[back to top]

  • An Overview of General Performance Metrics of Binary Classifier Systems [PDF]
  • Cross-validation
    • Streamline your cross-validation workflow - scikit-learn's Pipeline in action [IPython nb]



Parameter Estimation

[back to top]

  • Parametric Techniques

    • Introduction to the Maximum Likelihood Estimate (MLE) [IPython nb]
    • How to calculate Maximum Likelihood Estimates (MLE) for different distributions [IPython nb]
  • Non-Parametric Techniques

    • Kernel density estimation via the Parzen-window technique [IPython nb]
    • The K-Nearest Neighbor (KNN) technique
  • Regression Analysis

    • Linear Regression

    • Non-Linear Regression




Machine Learning Algorithms and Classification Models

[back to top]




Clustering

[back to top]

  • Protoype-based clustering
  • Hierarchical clustering
    • Complete-Linkage Clustering and Heatmaps in Python [IPython nb]
  • Density-based clustering
  • Graph-based clustering
  • Probabilistic-based clustering



Statistical Pattern Classification Examples

[back to top]

  • Supervised Learning

    • Parametric Techniques

      • Univariate Normal Density

        • Ex1: 2-classes, equal variances, equal priors [IPython nb]
        • Ex2: 2-classes, different variances, equal priors [IPython nb]
        • Ex3: 2-classes, equal variances, different priors [IPython nb]
        • Ex4: 2-classes, different variances, different priors, loss function [IPython nb]
        • Ex5: 2-classes, different variances, equal priors, loss function, cauchy distr. [IPython nb]
      • Multivariate Normal Density

        • Ex5: 2-classes, different variances, equal priors, loss function [IPython nb]
        • Ex7: 2-classes, equal variances, equal priors [IPython nb]
    • Non-Parametric Techniques




Resources

[back to top]

  • Matplotlib examples - Visualization techniques for exploratory data analysis [IPython nb]

  • Copy-and-paste ready LaTex equations [Markdown]

  • Open-source datasets [Markdown]

  • Free Machine Learning eBooks [Markdown]

  • Terms in data science defined in less than 50 words [Markdown]

  • Useful libraries for data science in Python [Markdown]

  • General Tips and Advices [Markdown]

  • A matrix cheatsheat for Python, R, Julia, and MATLAB [HTML]

About

A collection of tutorials and examples for solving and understanding machine learning and pattern classification tasks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • TeX 91.9%
  • CSS 8.1%