I want to serve my chatbot

nlp
1

Hi, I want to serve my chatbot. How can I do it? I have React-Node app, I think I will make an API (with Flask) and then connect with my app. Which ways should I use to build the Flask API? I trained my model on Colab. PyTorch Chatbot Tutorial

2

It’s good to see you’re wanting to productionize a model!

There’s now a tutorial about writing an API around your model using Flask. Then your react-node app would simply call it. Bearing in mind, that there’s probably a bit more you want to do before deploying to production, eg serve the Flask app with gunicorn or something similar.

Another approach would probably be to look into ONNX - I don’t know too much about that. I typically deploy models using Django or Flask.

1 Like