Defining api routes

[hieuthehungry]

I've recently picked up on using this framework. I can't find instruction on how to define the specific URL for the API, other than IP:port. What I want to do is deploying an API to an URL like IP:port/route/to/api. How do I achieve that? Thank you.

[lamhoangtung]

Hi @hieuthehungry. Thanks for the interest!

First of all, sorry for the lack of up to date documentation, since we are still iterating the library, we haven't really touch the documentation part.

For now, mlchain does not suport custom API route. There will be only the default route like host:port/call/method

For example, if your served model is something like this:

class Model:
    def predict(self, image):
        ....
        return res

    def predict_batch(self, images):
       ....
       return res

Running on 0.0.0.0 at port 9999. Then the server would only have 2 default route:

0.0.0.0:9999/call/predict
0.0.0.0:9999/call/predict_batch

I hope this can help ;)

[lamhoangtung]

Are the answer satisfied you @hieuthehungry. If it's not, feel free to let us know ;)

[hieuthehungry]

I got it, thanks for the answer.