[D] Neural nets. Hyper parameter manual optimization • r/MachineLearning

Does anyone know any paper that contains practical recommendations for the process of manual hyper parameter optimization? Such as which parameters to optimize first and which ones should be optimized in groups, etc. Thank you