Skip to content

Chatbot trained in a seq2seq model using Tensorflow

Notifications You must be signed in to change notification settings

Otazz/TF_Chatbot

Repository files navigation

Tensorflow Chatbot

Chatbot implementation using a trained in seq2seq model developed in Tensorflow and serving in a Flask server.

2-seq2seq-feed-previous Picture from Deep Learning for Chatbots

This project is divided in two parts: - The tensorflow implementation of the seq2seq model: chat.py - Webchat to get input from the user, "respond" them by running the model inference and display the output: main.py

The model construction is based on the following implementation Advanced dynamic seq2seq, with modifications involving new datasets and ways of training.

Front-end is using this simple chat implementation on https://bootsnipp.com/snippets/featured/simple-chat, modified to make requests to the Flask server.

About

Chatbot trained in a seq2seq model using Tensorflow

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published