The following tutorial introduces the use of an autoregressive model for generating sequences of dance poses. This model extends the model introduced in an earlier article<\/a> in that it combines a long short term memory (LSTM) network with a mixture density network (MDN). A MDN outputs the parameters for multiple gaussian distributions. In the autoregressive model employed here, each gaussian distribution can be sampled from to obtain one candidate pose for continuing a pose sequence. <\/p>\n\n\n\n This tutorial forms part of a series of tutorials on using PyTorch to create and train generative deep learning models. The code for these tutorials is available here<\/a>.<\/p>\n\n\n\n After 1000 epochs of training, the model generates pose sequences that look like this when rendered as skeleton animations.<\/p>\n\n\n\n