Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/lpfann/fri.jl
Simple proof of concept for feature relevance bounds written in Julia.
https://github.com/lpfann/fri.jl
feature-relevance feature-selection julia-package
Last synced: 26 days ago
JSON representation
Simple proof of concept for feature relevance bounds written in Julia.
- Host: GitHub
- URL: https://github.com/lpfann/fri.jl
- Owner: lpfann
- Created: 2019-11-12T15:24:04.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2020-02-09T12:21:24.000Z (over 4 years ago)
- Last Synced: 2024-10-10T05:35:35.018Z (27 days ago)
- Topics: feature-relevance, feature-selection, julia-package
- Language: Jupyter Notebook
- Size: 13.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fri.jl
This repository contains a simple [Julia](https://julialang.org) implementation of the feature relevance bounds method.
It exists a much more complete python version [here](https://github.com/lpfann/fri).
This is mostly a proof of concept and is missing features such as cross validation for hyper parameteres, regression and ordinal regression models and advanced statistics for feature classification.
# Quickstart
A runnable example is included in the example notebook.
```julia
include("src/Fri.jl")
```
┌ Info: Precompiling JuMP [4076af6c-e467-56ae-b986-b466b2749572]
└ @ Base loading.jl:1273
┌ Info: Precompiling ECOS [e2685f51-7e38-5353-a97d-a921fd2c8199]
└ @ Base loading.jl:1273
Main.Fri
We generate dataset with 200 samples, 5 strongly relevant features, 4 weakly relevant features and 10 noise features (irrelevant).
```julia
X,y = Main.Fri.dataset.generate(200,d_rel=5,d_irrel=10,d_weak=4);
```
```julia
relev_bounds = Main.Fri.relevance_bounds(X,y)
```
17×2 Array{Float64,2}:
2.55727 2.55727
2.01039 2.01039
2.32116 2.32116
2.09897 2.09897
2.35608 2.35608
-3.85029e-12 2.22748
-3.8504e-12 2.22748
0.0948441 0.0948441
0.100589 0.100589
0.156153 0.156153
0.0760626 0.0760626
0.11897 0.11897
0.19504 0.19504
0.09535 0.09535
0.0669823 0.0669823
0.22617 0.22617
0.0860516 0.0860516
## Minimal relevance for feature 1
```julia
relev_bounds[1,1]
```
2.5572730462733446
## Maximal relevance for feature 1
```julia
relev_bounds[1,2]
```
2.557273051045558