# Defining data transformations

- data transformation per column

We need to put type names for every column in the list:

example: [col1, col2] -> [col1 int, col2 string]

Our columns have to look like this:

objectid int
departmentid int
departmentname string
ishighlight boolean
accessionnumber string
accessionyear string
ispublidomain boolean
primaryimage string
constituents array
objectname string
title string
culture string
period string
dynasty string
reign string
objectdate string
objectbegindate int
objectenddate int
medium string
dimensions string
measurements array
geographytype string
city string
state string
country string
region string
classification string
metadatadate datetime
repository string
istimelinework boolean
gallerynumber string