LightML.jl

About

LightML.jl is a collection of reimplementation of general machine learning algorithm in Julia.

The purpose of this project is purely self-educational.

Why?

This project is targeting people who want to learn internals of ml algorithms or implement them from scratch.

The code is much easier to follow than the optimized libraries and easier to play with.

All algorithms are implemented in Julia.

You should access test function of every implementation for its usage in detail. Every model is actually constructed in a similar manner.

Installation

First make sure you have correct python dependency. You can use the Conda Julia package to install more Python packages, and import Conda to print the Conda.PYTHONDIR directory where python was installed. On GNU/Linux systems, PyCall will default to using the python program (if any) in your PATH.

The advantage of a Conda-based configuration is particularly compelling if you are installing PyCall in order to use packages like PyPlot.jl or SymPy.jl, as these can then automatically install their Python dependencies.

ENV["PYTHON"]=""
Pkg.add("Conda")
using Conda
Conda.add("python==2.7.13")
Conda.add("matplotlib")
Conda.add("scikit-learn")
Pkg.add("PyCall")
Pkg.build("PyCall")

or you can simply

Pkg.build("LightML")

It's actually same with the procedure above.

Then every dependency should be configured, you can simply run command below to install the package.

Pkg.clone("https://github.com/memoiry/LightML.jl")

Running Implementations

Let's first try the overall functionality test.

using LightML
test_LSC()

<p align="center"> <img src="https://ooo.0o0.ooo/2017/03/25/58d640c2c7a1a.png" width="640"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 1: Smiley, spirals, shapes and cassini Datasets using LSC(large scale spectral clustering) </p>

Running Demo

using LightML
demo()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//15//58c8cb6e1a1d3.png" width="640"></img> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 2: The Digit Dataset using Demo algorithms </p>

Current Implementations

Supervised Learning:

• Adaboost
• Decision Tree
• Gradient Boosting
• Gaussian Discriminant Analysis
• K Nearest Neighbors
• Linear Discriminant Analysis
• Linear Regression
• Logistic Regression
• Multilayer Perceptron
• Naive Bayes
• Ridge Regression
• Lasso Regression
• Support Vector Machine
• Hidden Markov Model
• Label propagation
• Random Forests
• XGBoost

Unsupervised Learning:

• Gaussian Mixture Model
• K-Means
• Principal Component Analysis
• Spectral Clustering
• Large Scale Spectral Clustering

Test Example available

• test_ClassificationTree()
• test_RegressionTree()
• test_label_propagation()
• test_LDA()
• test_naive()
• test_NeuralNetwork()
• test_svm()
• test_kmeans_random()
• test_PCA()
• test_Adaboost()
• test_BoostingTree()
• test_spec_cluster()
• test_LogisticRegression()
• test_LinearRegression()
• test_kneast_regression()
• test_kneast_classification()
• test_LSC()
• test_GaussianMixture() (Fixing)
• test_GDA() (Fixing)
• test_HMM() (Fixing)
• test_xgboost (Fixing)

Contribution

Please examine the todo list for contribution detials.

Any Pull request is welcome.

Selected Examples

LinearRegression

using LightML
test_LinearRegression()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//11//58c2cf6a8726e.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 3: The regression Dataset using LinearRegression </p>

Adaboost

test_Adaboost()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//11//58c36970c58a8.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 4: The classification Dataset using Adaboost </p>

SVM

test_svm()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//11//58c367760e76a.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 5: The classification Dataset using LinearRegression </p>

Classification Tree

test_ClassificationTree()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//11//58c36775113e6.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 6: The digit Dataset using Classification Tree </p>

kmeans

test_kmeans_random()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//02//18//58a8445e2114b.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 7: The blobs Dataset using k-means </p>

LDA

test_LDA()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//02//58b82861bade3.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 8: The classification Dataset using LDA </p>

PCA

test_PCA()

<p align="center"> <img src="https://ooo.0o0.ooo//2017//03//11//58c36773da5da.png" width="480"> </p> <p align="center" style="color:rgb(220,220,220);"> Figure 9: The Digit Dataset using PCA </p>