Getting labels for Unsupervised Dataset for finding the stressed word Sorry I didn't know how to question this well but stick with me. I have a dataset which contain 2 syllabic datasets,in which each word is either stressed which has label=1 or unstressed which has label=0 and i have 19 features for each example, and i have about 6000 examples, Now i was given y-labels which I used to train them with a simple DNN, and got a pretty good accuracy of about 89% , but now I have to train the model without the y-labels and i don't know where to start. Attempts I...