Hi, it my first time creating a post here.
I’m not sure if my project fits here. I hope I’ve done it right. Please let me know if anything I’ve done is not appropriate.
I built a machine learning deployment tool in my free time: here, if you have time, you can take a look.
Any feedback is welcome, thank you.
I’m trying to make this tool :
- to serve models much easier and straight forward.
- fast to prototype
- can serve any model.
- API compatible with Kubeflow, TF Serving, Triton and TorchServe. There is no pain switching to or from them.
- have a 100% test coverage so more robust.