being able to create a 'sequential' analysis would be cool #10
Description
Making neural networks in Keras is done by creating a model object and adding layers to it, like connecting machines on a conveyor belt (sort-of). Having something like this but for analytical pipelines in gamba would be really cool - not sure technically how to do this, but it would revolve around having some sort of abstraction capable of interacting with all parts of the analytic side of the library, and perhaps some parts of the data part.
This would be a very advanced feature to implement (I think?) so isn't a priority for now, but it would be nice eventually!