Doubt

[vishalsingha]

1. Is it necessary to use the same model can't we use other classification model?
2. Can we perform Hyperparameter tuning on the model we selected?
3. Also is it mendatory to size of train/size of cross_val set as mentioned in Notebook or we are free to choose by own

[goodrahstar]

@vishalsingha You need to use the same model, idea is to show the performance boost of the same model using better-engineered features.

[Cheril311]

@goodrahstar isn't even tuning of hyperparameters allowed in the same model?

[goodrahstar]

@Cheril311 You can do that, but more weightage is given on the feature engineering aspect.