I finally got it. so when you run the predict_proba on your test data you will get 5 columns and n rows so each row is a probablity of predicting that particular class. you only need to get the class name of your encoded target then map it. as columns name. that's all you need to do.
Same
I finally got it. so when you run the predict_proba on your test data you will get 5 columns and n rows so each row is a probablity of predicting that particular class. you only need to get the class name of your encoded target then map it. as columns name. that's all you need to do.
are you using fold predictions??
yes
Me too. Someone help