Method of determining a measurement recipe and associated metrology methods and apparatuses

A non-transitory computer-readable data carrier having stored thereon a computer program that, when executed by a computer system, is configured to cause the computer system to at least:

• obtain first measurement data relating to measurement of a plurality of reference targets, the reference targets comprising respective different overlay biases and being located in at least one reference target cluster for each exposure field on a training substrate, wherein the first measurement data further relates to measurement of the reference targets using a plurality of different acquisition settings for acquiring the measurement data;
• obtain second measurement data relating to measurement of a plurality of in-die targets, the in-die targets comprising respective different overlay biases and being distributed over each exposure field, wherein the second measurement data further relates to measurement of the in-die targets using the plurality of different acquisition settings for acquiring the measurement data;
• train one or more machine learning models using at least the first measurement data to obtain a plurality of candidate measurement recipes, wherein the candidate measurement recipes comprise a plurality of combinations of a trained machine learned model and a corresponding acquisition setting;
• determine a selected measurement recipe from the candidate measurement recipes using the second measurement data;
• and
• cause physical configuration an in-die target measurement process based on the selected measurement recipe and/or provide a signal representing, or based on, the selected measurement recipe to a system for use in configuration of the in-die target measurement process.