Which option describes a best practice when cleaning data?

Only fix errors that are found during a pipeline stage, and do not adjust the source data.

Always discard the raw data after cleaning to save valuable storage space.

Come to an agreement on what clean looks like.

Always make common$-$sense assumptions about what data value means.

aws

$$