The data schema is the meta information, the dictionary of a data file. It informs Amazon ML of the type of each variable in the dataset. Amazon ML will use that information to correctly read, interpret, analyze, and transform the data. Once created, the data schema can be saved and reused for other subsets of the same data. Although Amazon ML does a good job of guessing the nature of your dataset, it is always a good idea to double-check and sometimes make some necessary adjustments.
By default, Amazon ML assumes that the first row of your file contains an observation. In our case, the first row of the titanic_train.csv file contains the names of the variables. Be sure to confirm that this is the case by selecting ...
