What are the ways to deploy a model made in R. And how does it work in future (I. E If we deployed it so. Do. We jsut add the data evry month or need to. Rebuild it again and. Again)?
To deploy any machine learning model, I would prefer to use languages that are reliable in those cases, such as Java or Python. You can keep any machine learning model that you created as an API for standalone predictive systems and then take in the predictions in another system which is used to serving the predictions.
Now as the model is deployed, you can keep this model as it is or changes it as per business requirements. For example, if the nature of your data changes along with time, it is appropriate to keep your model updated along with those changes. This can be monthly / weekly or daily
Best way is to offer it via an API. You build your model and then using the “PLUMBER” package in R, turn it into API. If you’re using Python, you can use “FLASK”.
More details at : PLUMBER API in R