pythondriverless-ai

Does a mojo-pipeline automatically drop columns as specified during training?


I have ran a DAI experiment & built the mojo-pipeline. Now I have brand new data with columns I originally dropped within the DAI GUI; ID column for example. Does the mojo-pipeline automatically drop these columns when scored or do I need to do this prior?


Solution

  • The mojo-pipeline will ignore all columns that weren't used for training - that includes columns that were dropped in DAI and any new columns they may have been introduced into the data.