Timeline for Decision Trees vs Boosting, Random Forests
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 11, 2020 at 15:35 | vote | accept | Ged | ||
| May 11, 2020 at 15:35 | comment | added | Ged | That said, computing is cheap. | |
| May 11, 2020 at 15:05 | comment | added | Deepak Chaudhary | Yes they are used by and large because obviously they will perform better then base model of decision tree. But you should always consider the cost at which you achieved this increased in performance. Suppose you have only got 1% increase in performance and used good number of hours to train the model then it is not necessary to use AdaBoost. But if you only care about the performance then you should definitely use random forest or AdaBoost! | |
| May 11, 2020 at 8:07 | comment | added | Ged | So what do you think? | |
| May 10, 2020 at 20:20 | comment | added | Ged | OK. Not up to RF yet. But, I got that issue on shallow decision tree (hb, lv) from the course --> stumps. I then look at your 2nd para & 3rd para and they are correct but a little non-committal in terms of the question. The lecturer stated these boosting approaches are used by and large. So, my question is not entirely answered . | |
| May 10, 2020 at 20:08 | history | answered | Deepak Chaudhary | CC BY-SA 4.0 |