I'm working with a time series data in which each sample itself is a vector. In following plots, x is time step. The first two plots show positive example, and the other two are negative ones. My current plan is to apply slide window of fixed width to the series, and feed each window (now can be considered as a 2-D matrix) into a neural network. But I wonder if there're potentially less computationally intensive approaches.

enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.