Seonglae ChoTest Set estimate real-world performance
No decisions should be made using the test set
For unbiased estimate, test data should be used as a metric since they are unseen dataset.
The test set is used to evaluate the model's final performance and is not involved in the training process. In real-world applications, the test distribution often does not match the training distribution.
Adversarial Validation
Santiago on Twitter / X
I want to show you a clever trick you didn't know before.Imagine you have six months' worth of data. You want to build a model, so you take the first five months to train it. Then, you use the last month to test it.This is a common approach for building machine learning… pic.twitter.com/SaXfNK5HHu— Santiago (@svpino) April 2, 2024
https://twitter.com/svpino/status/1775154270708396215