What do you mean exactly? Is it that you want more visibility into what kind of preprocessing code is running for each endpoint?
Like Nathan said, custom engines are a TODO, but for your second question, you can add that API request in the model preprocessing, which is a function you can define yourself! It will be ran every time a request comes in and you can do whatever you want in it and change the incoming data however you wish 🙂
example: https://github.com/allegroai/clearml-serving/blob/main/examples/keras/preprocess.py
Because I want different models versions in the same API fashion. Some of them are RuleBased som of them are MLbased.
- At the inference I need to ask som other features from distinct sources. Ex: Some third ApI request.
Hi Oriel!
If you want to only serve an if-else model, why do you want to use clearml-serving for that? What do you mean by "online featurer"?
Hi DeterminedCrocodile36 ,
To use a custom engine you need to change the process tree.
https://github.com/allegroai/clearml-serving/tree/main/examples/pipeline
Section 3 is what you're interested in
And here is an example of the code you need to change. I think it's fairly straightforward.
https://github.com/allegroai/clearml-serving/blob/main/examples/pipeline/preprocess.py
But if I'm going to use Preprocessing Layer for tha then I won't be able to seeat ClearML Models the methods for RuleBased vs The ML based models.
Adding a custom engine example is on the 'to do' list but if you manage to add a PR with an example it would be great 🙂