Need help with pytorch_chatbot?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

jinfagang
220 Stars 54 Forks 3 Commits 3 Opened issues

Description

A Marvelous ChatBot implement using PyTorch.

Services available

!
?

Need anything else?

Contributors list

No Data

PyTorch Marvelous ChatBot

[Update]

it's 2019 now, previously model can not catch up state-of-art now. So we just move towards the future a transformer based chatbot, it's much more accurate and flexiable as well as full of imagination interms of being a cahtbot!!! More importantly we are opensourced the whole codes here: http://manaai.cn/aicodes_detail3.html?id=36 Be sure to check it if you interested in chatbot and NLP!! it's build with tensorflow 2.0 newest api!!

Aim to build a Marvelous ChatBot

Synopsis

This is the first and the only opensource of ChatBot, I will continues update this repo personally, aim to build a intelligent ChatBot, as the next version of Jarvis.

This repo will maintain to build a Marvelous ChatBot based on PyTorch, welcome star and submit PR.

Already Done

Currently this repo did those work:

  • based on official tutorial, this repo will move on develop a seq2seq chatbot, QA system;
  • re-constructed whole project, separate mess code into
    data
    ,
    model
    ,
    train logic
    ;
  • model can be save into local, and reload from previous saved dir, which is lack in official tutorial;
  • just replace the dataset you can train your own data!

Last but not least, this project will maintain or move on other repo in the future but we will continue implement a practical seq2seq based project to build anything you want: Translate Machine, ChatBot, QA System... anything you want.

Requirements

PyTorch
python3
Ubuntu Any Version
Both CPU and GPU can works

Usage

Before dive into this repo, you want to glance the whole structure, we have these setups:

  • config
    : contains the config params, which is global in this project, you can change a global param here;
  • datasets
    : contains data and dataloader, using your own dataset, you should implement your own dataloader but just a liitle change on this one;
  • models
    : contains seq2seq model definition;
  • utils
    : this folder is very helpful, it contains some code may help you get out of anoying things, such as save model, or catch KeyboardInterrupt exception or load previous model, all can be done in here.

to train model is also straightforward, just type:

python3 train.py

Contribute

wecome submit PR!!!! Let's build ChatBot together!

Contact

if you have anyquestion, you can find me via wechat

jintianiloveu

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.