The file is not located in local machine to use URL(fileURLWithPath: dataPath), it has a URL link. In addition to that the Header is on 3rd row so once you change hasHeaderRow: true then it crashes. So couldn't load the header from 3rd row. Would you mind running the avove code with hasHeaderRow: true

ParsingOptions() in MLDataTable has skipRows option where as CSVReadingOptions in DataFrame lacks the skipRows