The Chatbot : Another Foray into the Vast World of Machine Learning

Confession : This is me trying to dig myself out of tutorial hell. I've lost count of the number of times I've wanted to try something - then start watching a tutorial - lose the plot halfway there - only to blindly follow along and type out commands wanting to look at a final product and feel like I've accomplished something. And that is rarely ever the case. Sound familiar? It's a pretty common problem I hear.

So what am I doing here? Well right now I find myself wanting to be able to apply TensorFlow in my future projects. In order to achieve this hefty goal, I will be building a fully end-to-end Chatbot application that responds to user input. You can check out the progress I make here:

JDPrabasha / chatbot

A chat application that allows you to chat with an AI-powered chatbot

Chatbot Application

This repository contains code for a chatbot application that responds to user input, and is based on the YouTube tutorial by NeuralNine, which you can find here : https://www.youtube.com/watch?v=1lwddP0KUEg

Prerequisites

• nltk
• TensorFlow
• Numpy
• Flask

How to Run

Run chatbot.py to start the flask server

Then cd into the Client folder and run

npm install

to install the required dependancies.

Then run

npm start

to access the application on your local browser

View on GitHub

This is not the first time I will be using TensorFlow. Nor is it the first time I will be building a Web Application. Will I be following tutorials? Absolutely.

The catch - this time I will be documenting all of my thought processes, problems faced, lessons learned and all those other things in-between. Essentially, if all goes to plan, this series of articles should allow anyone to speedrun their way through to achieving a similar goal. Why do it this way you ask? Well, if I can explain that entire process, that should mean all those things must have stuck with me yeah? Fingers crossed 🤞

Here's the current plan :

1. Follow along with this tutorial : and build the model for the application.
2. Connect the model to a Flask backend
3. Connect the Flask application to a React frontend
4. Add styling

And well that's all I can think of so far I'm afraid. I'll be wrapping up things here for now. Wish me luck!