Feature Technology
csv` desk, and i also started to Bing many things eg “Just how to profit a great Kaggle race”. All of the performance said that the answer to effective was feature engineering. Thus, I decided to feature engineer, but since i don’t really know Python I can not perform they towards the shell off Oliver, so i returned to kxx’s code. I feature designed particular articles according to Shanth’s kernel (We hands-composed aside all categories. ) up coming provided it to your xgboost. They had local Cv off 0.772, along with social Lb regarding 0.768 and private Pound from 0.773. Very, my ability technology did not assist. Darn! To date I wasn’t thus reliable out-of xgboost, thus i attempted to rewrite the fresh new code to utilize `glmnet` playing with collection `caret`, however, I didn’t learn how to boost a mistake We had while using `tidyverse`, thus i eliminated. You will find my code of the pressing right here.
On may twenty-seven-30 We returned so you can Olivier’s kernel, but I ran across that we didn’t only only have to carry out the suggest toward historic tables. Continue reading “Following this, We watched Shanth’s kernel on carrying out additional features about `bureau”