https://github.com/hmjianggatech/esmote
ESmote - An R package implemneting fast SMOTE algorithm
https://github.com/hmjianggatech/esmote
imbalanced-learning machine-learning r smote
Last synced: about 1 month ago
JSON representation
ESmote - An R package implemneting fast SMOTE algorithm
- Host: GitHub
- URL: https://github.com/hmjianggatech/esmote
- Owner: HMJiangGatech
- Created: 2017-06-28T15:58:22.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2019-10-30T00:13:14.000Z (over 5 years ago)
- Last Synced: 2025-01-31T16:49:23.966Z (3 months ago)
- Topics: imbalanced-learning, machine-learning, r, smote
- Language: C++
- Homepage:
- Size: 13.1 MB
- Stars: 4
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# esmote
**`esmote`**, an R package including fast SMOTE algorithm.
This is part of my undergraduate final year project. Which provide a really fast implementation of SMOTE algorithm.
If you have any concerns please contact me: [email protected]
Some functions are still underconstruction.
Ex. I developed a semi-supervised autoencoder to deal with high-dimensional data. However I did not provide a well documented R warpper. You may refer to the source code (`./tests/testPer.R`).
## Installation
First install `devtools` in `R`:
```R
install.packages("devtools")
```
Install package via `install_github`:
```R
library(devtools)
install_github('HMJiangGatech/ESmote')
```
## Practical Example
This package contains some test data, such as hand written digits data.
```R
newlabel = digitsTrainLabel;
newlabel[newlabel>0] = 1;
newID = sample(60000);
timestart<-Sys.time();
newdata<-esmote::Smote(digitsTrain[newID,],newlabel[newID], algorithm="rp_forest");
timeend<-Sys.time()
runningtime<-timeend-timestart
print(runningtime)
```
Compared to other packasges such as: `DMwR`, `smotefamily`, it is extremely fast.