Query regarding accuracy

#1

I submitted titanic survivor prediction on kaggle, I used random forest classier to train & predict. Accuracy score which I got is 77.2%. So is 77% decent accuracy for random forest classifier ? what maximum accuracy we can get using random forest classifier ?

#2

Hey @pasta 75% - 80% is nearly the maximum achievable accuracy with random forest. But in case if you still want to achieve more than you need to switch to DL techniques. But for the time being no need to ponder on that, and proceed further with the course.

Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section. :blush:

#3

You try your hyperparameters,min_samples_leaf=3,n_estimators=100,max_features=0.5