PyPI version

easyLDA is a library that easily build LDA Topic Models with just a list of docs (e.g. a list of twitter posts in CSV/TXT)

github: https://github.com/shichaoji/easyLDA

The library pipeline text preprocessing, such as tf-idf, n-grams from Gensim library

Credit to:




$ pip install easyLDA

usage example

simple need a text file (.csv) with each row represents a document (a post, comment, short article etc.), with only one column which is the text

text file (csv) sample view

Demo 1

easy to use, just in a shell window, type: easyLDA, then specify the location of the text document

1. then choose how many topics you want the model to fit

2. choose the topic contains only single word (1) or can be phases (2/3) as well

the program will be starting to train

Demo 2

model result

models folder created by program contains the trained model

xx.html file is the interactive visulization of the model result

Demo 3

visualization live example


static pic result

Demo 4