This is an example service using gRPC, machine learning, and a bunch of other cool tech.
To find more details, please look at my blog!
All you need is Go (I think)!
To run this repo, run make, this will create a model theta.bin (naming is hard), and copy that model over to both the predict cmd and the predict svc.
After that, if you run either go run cmd/predict/main.go to run the model against some currently hardcoded values
or
Better yet, you can run go run svc/main.go this will spin up an RPC service, where you can call Predict with the 4 values for sepal/petal width/length using a tool like evans or grpcurl.
The utilities in cmd are used to play and experiment with the data outside of the gRPC service.
To run the training utility;
go run cmd/train/main.go
This spits out a classifier model.cls, which we can then use in the prediction utility;
go run cmd/predict/main.go