Method and apparatus for performing vision test procedure on person

The invention relates to a computer-implemented method for training (500) a machine learning model (200) to determine a confidence value during at least one test loop (204) of a vision test procedure, where the confidence value is specified for determining at least one action in at least one subsequent test loop (204) of the vision test procedure. The invention also relates to a trained machine learning model (200), a computer program comprising instructions for training a machine learning model (200), and a training device (600). Furthermore, the invention relates to a computer-implemented method for performing a vision test program (202) on a person (300), a computer program comprising instructions for performing a vision test program, a vision test device (100), and a method (704) for generating a geometric model of at least one spectacle lens for manufacturing at least one spectacle lens. 本发明涉及一种用于训练(500)机器学习模型(200)以在视力测试程序的至少一个测试循环(204)期间确定置信度值的计算机实现的方法，其中，该置信度值被指定用于确定视力测试程序的至少一个后续测试循环(204)中的至少一个动作。本发明还涉