search

epsilon-machine / missingpy

Missing Data Imputation for Python
GNU General Public License v3.0
236 stars 63 forks source link

Miss Forest #30

Open CreaperLost opened 3 years ago

CreaperLost commented 3 years ago

Miss Forest doesn't train on the data. It just iteratively fit and applies RF trees on data. But on transform is should use the trained trees during fit no redo training on test data.

JStuckner commented 2 years ago

Completely agree. I wasted half a day figuring out that it retrains on the test data and is therefore useless to me. I was hoping to make predictions on test data with missing values.