Table of Contents
About category
This section covers various topics on Recurrent Neural Networks architectures and their usage.
While Convolutional networks (CNNs) seem to be very intuitive, some people have a hard time understanding the LSTM, GRU, or attention mechanisms. Posts in this category will try to explain at least some of the more complex concepts.
Posts in this category
-
Fifty Shades of Time: Time series & RNNs
Time series processing is a well-known and extensively researched problem. Moreover - many people naturally understand, what a time series is, and how it is structured. Things start to get complicated once you are about to prepare time-dependent data for modeling with neural networks. Three-dimensional structures are not that easy to imagine initially.