Add HTTP prediction server to the CLI

modelfoxdotdev / modelfox

ModelFox makes it easy to train, deploy, and monitor machine learning models.

Other

1.46k stars 63 forks source link

Add HTTP prediction server to the CLI #17

Closed nitsky closed 2 years ago

nitsky commented 3 years ago

Right now it is possible to make predictions from the CLI over stdin/stdout with tangram predict, but it would be nice to add an HTTP server for serving predictions which could be started like so: tangram serve --model <MODEL_PATH>.