hmc-whisk / jsLDA

A React based version of jsLDA with brand new features added on
Other
0 stars 0 forks source link
javascript lda natural-language-processing reactjs topic-modeling

jsLDA

An implementation of latent Dirichlet allocation in javascript by WHISK Lab at Harvey Mudd College. This version is an adaptation of David Mimno's original jsLDA.

Status

This repository is being actively developed and maintained.

Installation

First, clone the repo

git clone git@github.com:hmc-whisk/jsLDA.git
cd jsLDA

Then, install all dependencies

yarn

# using npm (not recommended)
npm i

If you don't have yarn installed but do have npm, you can install it by npm i -g yarn. yarn is much faster than npm.

Running the code

To run a development version of the code locally

yarn start

# or, for npm 
npm run start

To build a release version of the code

yarn build

# or, for npm
npm run build

Deploying the code

If you have access to a server under cs.hmc.edu, you can deploy a copy of the code using the deploy script (requires bash and rsync, so if you are using Windows you probably should run this under WSL)

yarn deploy <your-hmc-username>

# or, for npm
npm run deploy <your-hmc-username>

The code will then be live under www.cs.hmc.edu/~your-hmc-username/jsLDA/.