Deep metric learning to find a close era to live

It would be great if we could know which training or validation era is ‘close’ to the live era to enhance our model performance in the tournament. This knowledge may help us avoid a burning era by focusing our modeling on improving its validation performance on that ‘close’ era.

Here I use a simple deep metric learning approach to do that in the following kaggle notebook.

metric learning and live era

Feel free to comment and upvote if you like:)


This is pretty cool. All of these explanatory style posts / notebooks are awesome for beginners like myself. Thank you.

One of those final plots looks eerily like some meteorological image :leaves:.