Does anybody have an idea how one could convert a model coded in DecisionTree.jl into a TensorFlow model
The reason I would want that is e.g. because Firebase can host TensorFlow Lite models (converted from TensorFlow models) and it would be awesome to host a DecisionTree model with Firebase
I see that DecisionTree models can be stored as .jld2 (is this actually correct?). Can .jld2 model be converted to e.g. .GraphDef for TensorFlow?
Or any other ideas?