'When we use XGBOOST in binary classification, do we need to MinMaxscale the features we input?
Given the retweet dynamic of a tweet for a certain period T, we want to predict whether the tweet will be widely spread?
The model we selected is XGBOOST
I have got about 10 dimension features. If I conduct the process of Minmaxscale (or other standard methods), the performance of the model can be improved. If I do not conduct the above process, the performance of the model can not be improved.
However, many people told me that there is no need to do this in the model xgboost
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|