https://github.com/mlampros/fasttext

R package for 'Efficient Learning of Word Representations and Sentence Classification'
https://github.com/mlampros/fasttext

cpp11 fasttext r

Last synced: 8 months ago
JSON representation

R package for 'Efficient Learning of Word Representations and Sentence Classification'

• Host: GitHub
• URL: https://github.com/mlampros/fasttext
• Owner: mlampros
• License: other
• Created: 2019-04-11T19:13:54.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2024-02-15T10:18:34.000Z (over 2 years ago)
• Last Synced: 2025-04-11T01:49:11.462Z (about 1 year ago)
• Topics: cpp11, fasttext, r
• Language: C++
• Homepage: https://mlampros.github.io/fastText/
• Size: 3.34 MB
• Stars: 42
• Watchers: 3
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.Rmd
  • License: LICENSE

Awesome Lists containing this project

README

          

---

output: github_document

---

[![tic](https://github.com/mlampros/fastText/workflows/tic/badge.svg?branch=master)](https://github.com/mlampros/fastText/actions)

[![codecov.io](https://codecov.io/github/mlampros/fastText/coverage.svg?branch=master)](https://codecov.io/github/mlampros/fastText?branch=master)

[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/fastText)](http://cran.r-project.org/package=fastText)

[![Downloads](http://cranlogs.r-pkg.org/badges/grand-total/fastText?color=blue)](http://www.r-pkg.org/pkg/fastText)



[![Dependencies](https://tinyverse.netlify.com/badge/fastText)](https://cran.r-project.org/package=fastText)

## fastText




The **fastText** R package is an interface to the [fastText](https://github.com/facebookresearch/fastText) library for efficient learning of word representations and sentence classification. More details on the functionality of fastText can be found in the 

* [fastText_updated_version](http://mlampros.github.io/2019/04/11/fastText_updated_version/) (blog post)

* [fasttext_language_identification](http://mlampros.github.io/2021/05/14/fasttext_language_identification/) (blog post)

* [package documentation](https://mlampros.github.io/fastText/reference/index.html).




The [official website of the fasttext algorithm](https://fasttext.cc/) includes more details regarding the supervised & unsupervised functions. The following image shows the difference between [**cbow** and **skipgram**](https://fasttext.cc/docs/en/unsupervised-tutorial.html#advanced-readers-skipgram-versus-cbow) (*models to compute word representations*)




![](./man/figures/skipgram_vs_cbow.png)




Moreover, the following figure - extracted from [a survey (scientific paper) related to word embeddings](https://hal.science/hal-03148517/document) and recent advancements in Large Language Models - shows the differences between *static* and *contextualized* word embeddings

 


![](./man/figures/static_contextualised_word_embeddings.png)




You can either install the package from CRAN using,

```R

install.packages("fastText")

 

```




or from Github using the *install_github* function of the *remotes* package,

```R

remotes::install_github('mlampros/fastText')

```




**or** directly download the fastText-zip file using the **Clone or download** button in the [repository page](https://github.com/mlampros/fastText), extract it locally (rename it to *fastText* if necessary and check that files such as DESCRIPTION, NAMESPACE etc. are present when you open the fastText folder) and then run,

```R

#-------------

# on a Unix OS

#-------------

setwd('/your_folder/fastText/')

Rcpp::compileAttributes(verbose = TRUE)

setwd('/your_folder/')

system("R CMD build fastText")

system("R CMD INSTALL fastText_1.0.1.tar.gz")

#------------------

# on the Windows OS  

#------------------

setwd('C:/your_folder/fastText/')

Rcpp::compileAttributes(verbose = TRUE)

setwd('C:/your_folder/')

system("R CMD build fastText")

system("R CMD INSTALL fastText_1.0.1.tar.gz")

```




Use the following link to report bugs/issues (for the R package port),





[https://github.com/mlampros/fastText/issues](https://github.com/mlampros/fastText/issues)




### **Citation:**

If you use the **fastText** R package in your paper or research please cite both **fastText** and the **original articles / software** `https://CRAN.R-project.org/package=fastText`:




```R

@Manual{,

  title = {{fastText}: Efficient Learning of Word Representations and

    Sentence Classification using R},

  author = {Lampros Mouselimis},

  year = {2021},

  note = {R package version 1.0.3},

  url = {https://CRAN.R-project.org/package=fastText},

}

```