Deploying my rasa chatbot on my live website is my next step. While Rasa worked smoothly on my localhost server, as a newcomer to web development, I found the official guide provided by RASA in the link below a bit challenging to comprehend:
The RASA guide instructs me to include the following content in a credentials.yml file:
socketio:
user_message_evt: user_uttered
bot_message_evt: bot_uttered
session_persistence: true
Furthermore, it mentions the option of using REST Channels with a script command: run rasa. However, there's no indication if I need to transfer the project folder to my website or how to replicate the VENV setup, or even how to execute the "rasa run" script from my site.
This is what I accomplished in my local setup:
- I edited the credentials.yml with the code snippet stated above along with: socketio: user_message_evt: user_uttered bot_message_evt: bot_uttered session_persistence: true rasa: url: "http://localhost:5002/api"
- I activated the Venv named vvv via: conda activate vvv
- I ran the command:
(inside the project's directory)rasa run -m models --enable-api --cors "*" --debug
https://i.sstatic.net/aCVDE.png
- In my index.html file, the value of socketUrl was set to "http://localhost:5005"
https://i.sstatic.net/qQSjA.png
As shown in the screenshots below, the chatbot was functioning well:
https://i.sstatic.net/7rWE4.png
My challenge now is figuring out how to deploy this on my own website. Attempting to directly copy the folder to my website didn't yield any results. How can I trigger the command "rasa run -m models --enable-api --cors "*" --debug" from my website? Is it just a matter of integrating a simple JavaScript code to call upon the assistant, or does it involve more complex steps?