chadng / siml

Machine Learning algorithms implemented from scratch

Synopsis

This repository contains popular Machine Learning algorithms, which have been introduced in various blog posts (http://ataspinar.com). Most of the algorithms are accompanied with blog-posts in which I try to explain the mathematics behind and the interpretation of these algorithms.

Motivation

Machine Learning is fun! But more importantly, Machine Learning is easy. But the academic literature or even (wikipedia-pages) is full with unnecessary complicated terminology, notation and formulae. This gives people the idea that these ML algorithms can only be understood with a full understanding of advanced math and statistics. Stripped from all of these superfluous language we are left with simple maths which can be expressed in a few lines of code.

Notebooks explaining the mathematics

I have also provided some notebooks, explaining the mathematics of some Machine Learning algorithms.

• Linear Regression and Logistic Regression

• Naive Bayes Classification

• Perceptron Classification

• Classification with Scikit-Learn

• Machine Learning with Signal Analysis techniques

Notebooks explaining Machine Learning with the Wavelet Transform

• Introduction to PyWavelets (for Wavelet Analysis
• Using Wavelets to Visualize the Scaleogram, time-axis and Fourier Transform
• Classification of signals using the Continuous Wavelet Transform and Convolutional Neural Networks
• Classification of ECG signals using the Discrete Wavelet Transform and Gradient Boosting
• Classification of signals using the Discrete Wavelet Transform and several classifiers

Installation

To install siML:

(sudo) pip install siml

or you can clone the repository and in the folder containing setup.py

python setup.py install

Code Example

TODO