Titanic surviver

#1

sir iss prject me 81% accuracy se jyada achieve nhi ho pa rhi
i tried with diff parameters like depth min_sample_split etc
in ensemble model with diff base learners i got 81 % so what should i do?

#2

hey @Ashu1318 ,
Only changing model parameters and taking other models are not the only thing to be done for improvement.

Try in this manner.

  1. Feature engineering - : Understand the data , check collinearity , create new features , or encode the given one.
  2. Modelling : With default parameters try different models and check how they perform.
  3. Hyperparameter tuning, once you select a model for this , then start tuning it.

It will be more beneficial and useful.

#3

okay sir look i got the feature which are highly corrleated so how can we genarate new features if we randomly add/sub features it would consumes lots of time

#4

is there any shortcut in sklearn which automatically generate new feature

#5

or how should we think or procede after looking feature_importances of data

#6

there are multiple way to generate those features but it all depends upon the data.
SO you have to work accordingly.

#7

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.