Time Series with PyTorch

15

Time Series Clustering

“If intelligence were a cake, unsupervised learning would be the cake, supervised learning the icing, and reinforcement learning the cherry on top. We know how to make the icing and the cherry, but we don’t yet know how to bake the cake.” – Yann LeCun

In the previous chapter, we focused on classification methodologies, where time series data labelled with predefined categories is used to train models through supervised learning (implemented via Aeon). A critical aspect of this approach involved providing algorithms with direct or transformed time series data relative to known labels. Here we build on the idea of modelling transformed time series to discover shared patterns that bind temporal data together, in an unsupervised ...