Code for the paper "Incrementally Updated Spectral Embeddings"
The scripts use 3 different datasets: college-msg
is available from the SNAP dataset
collection here, while the
temporal-reddit-reply
dataset can be found here. Finally, the household power consumption dataset has been downloaded from the UCI ML repository, and is available here.
All 3 datasets should be downloaded and stored under a folder called
datasets/
(which the scripts point to by default). Alternatively, the
--filepath
option should be used to override the location of each data file,
when applicable.