The project structure

Note the Rasa framework must be installed

Must be using Python version 3.8

The project is split into 3 parts:
Frontend
Backend
Chatbot

The front end can be run by navigating to the frontend using

cd ./frontend
npm i
npm start

The chatbot must be run using two terminals and can be run using

cd ./rasa

in one terminal run:

rasa run actions

in another terminal run

rasa train
rasa run --enable-api --cors="*" --credentials credentials.yml

The backend does not need to be run

The backend is currently being hosted at: http://unn-w19007452.newnumyspace.co.uk/kv6003/api

React files explained:

inside ./frontend/src/ the assets folder contains icons gathered from https://material.io/design/iconography/system-icons.html#design-principles
The css folder and App.css contain all the css for the frontend. The components folder is split into 5 subfolders.
Forms are the forms displayed on the modals.
Modals are the popup windows such as login and settings.
Pages include the main homepage and an error page.
Sections includes the split up sections to be displayed on the homepage.
ui_elements are the interface elements displayed on the homepage.

Rasa files explained:

inside ./rasa the config.yml file contains the pipeline for the rasa model including tokenizers, featurizers etc. Credentials includes additional endpoints to be called by the frontend (endpoints are found in the actions folder). The domain.yml file contains a reference to all intents and responses.

Inside the ./rasa/data/ folder the nlu.yml file contains all the intent training data. The rules contains rules which must be followed when an intent is recognised. The stories file contains training data indicating which action should be taken when an intent is recognised.

Inside the ./rasa/actions/ folder the actions.py file contains all the customised responses, each response must be a class and have a run method. The remaining files communicate with the API and contain methods related to their respective names. Inside the constants folder is a constants.py file containing all the constants for the actions.py file. The custom actions folder contains customer actions to be called by the front end, these include writing to files, retraining models and changing the current models.