30 sats \ 1 reply \ @random_ 23 May \ parent \ on: Amboss Targets Bitcoin Scalability with New Machine Learning Research lightning
Group by channel ID and then split.
https://dplyr.tidyverse.org/reference/group_split.html
That leaks node-level features from the train set into the test/validation set then no? Like if 90% of my channels are in the train set, and the top performing feature is "positional encoding" then the train set can basically learn my nodes positional encoding and whether my channels tend to be balanced or not. Then it just extrapolates that to my channels in the test/validation set.
reply