System and method for generating SEM quality metrology data from optical metrology data with machine learning

In one embodiment, one or more non-transitory machine-readable media have instructions thereon that, when executed by a processor, are configured to perform operations that include obtaining scanning electron microscope (SEM) metrology data for a first region on a training wafer, obtaining optical metrology data for a second region on the training wafer, and obtaining scanning electron microscope (SEM) metrology data for a third region on the training wafer. And training a model by using the SEM metrology data and the optical metrology data of the training wafer to generate a parameter of the feature on the production wafer based on the optical metrology data for the region of the production wafer. 在一个实施例中，一种或多种非暂态机器可读介质上具有指令，当处理器执行指令时，指令被配置为执行操作，该操作包括：获取训练晶片上的第一区域的扫描电子显微镜(SEM)量测数据，获取训练晶片上的第二区域的光学量测数据，以及通过使用训练晶片的SEM量测数据和光学量测数据来训练模型，以基于针对生产晶片的区域的光学量测数据来生成生产晶片上的特征的参数。