Project for open sourcing research efforts on Backward Compatibility in Machine Learning
71
stars
14
forks
source link
Raw values section (BTC, BEC, h1/h2 accuracy, etc.) should make it clear if point chosen is from training or testing set #129
Open
jlema opened 3 years ago
This could use something like an info bar under or above the label that explains 'data below is calculated from (testing / training) data.'