When should you add a 'Mislabeled' category to your error analysis spreadsheet?

True or False: You should add a 'Mislabeled' category to your error analysis spreadsheet only when you are certain—not merely suspicious—that examples are mislabeled.

When you suspect a significant fraction of dev set examples have incorrect labels, you should add a _____ column to your error analysis spreadsheet.

When should you add a 'Mislabeled' category to your error analysis spreadsheet according to Machine Learning Yearning?

The 'Mislabeled' column in the error analysis spreadsheet tracks examples where the ground-truth label itself is incorrect.

You should add a Mislabeled category to the error analysis spreadsheet when you suspect the _____ of mislabeled images is significant.

Match each column from the Machine Learning Yearning error analysis spreadsheet to what it captures.

Order the steps for setting up and using a Mislabeled column in an error analysis spreadsheet.

Which of the following is NOT listed as a column in the error analysis spreadsheet shown in Machine Learning Yearning (p. 33)?

According to Machine Learning Yearning, a 'Mislabeled' column should be included in every error analysis spreadsheet as a standard default.

The error analysis spreadsheet in Machine Learning Yearning has columns: Image, Dog, Great cat, Blurry, _____, and Comments.

Match each description to the corresponding column name in the Machine Learning Yearning error analysis spreadsheet.

Order the reasoning steps that lead a practitioner to add a 'Mislabeled' column rather than ignoring suspected label errors.

Explain how to set up error analysis when dataset labels are suspected to be incorrect.

Case study: Designing an error analysis spreadsheet for a dog classifier with incorrect labels.

Identify the tracking purpose and trigger for the 'Mislabeled' column.