When should we use a function to adjust learning rate with training proceeding?

hi I found that in some tutorials regarding training DNNs, they use learning rate adjusting function. But in some others there is nothing like that. Can anybody tell me when and why to use that function?

I don’t think there is a general rule which says how to set the learning rate. As far as I understand, this is something that you learn using trial and error to build up your intuition about the dataset and the network you use.