Use a model inside PerceptiLabs (Batch inference) | Voters

Use a model inside PerceptiLabs (Batch inference)

PerceptiLabs

Add the possibility to run data through a model inside the tool and get the results. This way there is no need to export the model and create an external tensorflow script just to run it.

Created by Robert Lundberg

August 13, 2021

Robert Lundberg

Hey all!

With the recent release of FastAPI and Gradio, do you still feel that this is needed more than just a "nice to have"? :)

Julian Moore

Robert Lundberg: Well, since PL has some nice metric visualisations, though it still leaves it as "nice to have" it would be very nice to be able to do this to compare model performance on arbitrary datasets for performance characterisations without having to re-code externally.

Robert Lundberg

Julian Moore: That makes sense, thanks!

Robert Lundberg

marked this post as

open