zshwuhan / Neural-Dynamic-Focused-Topic-Model

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Neural Dynamic Focused Topic Model

Code accompanying the paper Neural Dynamic Focused Topic Model.

Installation of the library

In order to set up the necessary environment:

  1. create an environment drf with the help of [conda],

    conda env create -f environment.yaml
    
  2. activate the new environment with

    conda activate drf
    
  3. install drf with:

    python setup.py install # or `develop`
    python setup.py install # or
    
    

Data Preprocessing and preparation

All the data used should be in the folder {project_location}/data. The preprocessing of the *NIPS, ACL and UN datasets is done using the script located in {project_location}/scripts/preprocessing/dataset_preprocessor.py.

Training of models

The training of the models used in the paper is done by using the training scripts located in the {project_location}/scripts/train/{acl|nips|un|nips_perrone} folder.

Evaluating of the trained models

For evaluation of the trained models we use the script {project_location}/scripts/evaluate/topic_models_evaluate_all.py

About

License:MIT License


Languages

Language:Python 99.6%Language:Shell 0.4%