The schema is a JSON file, which implies that we can create one from scratch for our data or modify the one generated by Amazon ML. We will now modify the one created by Amazon ML, save it to S3, and use it to create a new datasource that does not include the body, boat, and home.dest variables.
Click on View Input Schema as shown in the next screenshot:
This gives us the raw schema in JSON format. Save it on your local machine with the filename titanic_train.csv.schema. We will load this file on S3 in the same bucket/folder where the titanic_train.csv file resides. By adding .schema to the data CSV filename, we allow Amazon ...