I have data like this (column ObjectID and Cost) and I normally split data using:
sample.split(data$Cost, SplitRatio=0.7)
But in this case I want keep 'Cost' as relative ratio but also I want to use ObjectID - so each ObjectID can be only in test or train group. How to random split this?
ObjectID Cost Type
12345 1624 Test
12345 1175 Test
12345 1049 Test
12345 1733 Test
11111 1945 Train
11111 1989 Train
22222 1448 Test
22222 1815 Test
22222 1244 Test
33333 1355 Train
33333 1134 Train
44444 1478 Train
44444 1082 Train
44444 1147 Train
44444 1290 Train
55555 1383 Train
55555 1378 Train
55555 1288 Train
Aucun commentaire:
Enregistrer un commentaire