I am considering using Dataiku but I have one question left that I cannot find an answer for.
When training one or several models, the results are displayed in an experiment along with valuable metrics. I have seen that it is possible via a script to save the model file but is there also a way to save the experiment results as they are displayed in Dataiku in a more interactive way (something more elaborate than the chosen values in a text file) ?
Basically, I am wondering if it is possible to do something like the experiment feature in Azure ML which keeps tracks of every run with visualizations, so that it is possible to evaluate a model evolution through different runs. I didn't find something like that in the documentation but maybe you know some way to approach it.