Update with Archer's fix using multiple pipelines

[stephqwu]

To handle columns that require several transformations

This PR also ran code that updates data folders with the newest versions of our datasets