1
$\begingroup$

Background: Its well known that Pytorch and TensorFlow are currently the most used frameworks for Deep Learning (DL) research. As far as I know, most researchers (applied or theoretical) that contribute to the field of DL usually perform experiments with Pytorch. Specifically, the level of abstraction is just right to try custom architectures or models without having to build everything from scratch.

Question: What about research in another popular field of machine learning, tree-based methods and ensembles? I am thinking about ExtraTrees, XGBoost, LightBoost, CatBoost etc.

  • Are there any similar frameworks (like Pytorch or Tensorflow) that researchers use to experiment and develop new tree-based methods and ensembles?
  • Are Pytorch or Tensorflow suited for research in this field, or are they only for DL research?
  • There is also scikit-learn, but my guess is that it is too high level for tinkering and researching new methods.
  • If none of the above, is object oriented programming (from scratch) in e.g. Python or C++ the best way to perform research in this field?
Improve this question
$\endgroup$
1
  • 2
    $\begingroup$ I don't know but I suspect that the last option is much more practical if one wants to experiment with decision trees. Contrary to DL, implementing a full decision tree algorithm is fairly easy, and this gives maximum flexibility about how it works internally and how it can be used externally in ensemble methods.Also I suspect that there's little research left to be done in this domain, as opposed to DL. $\endgroup$
    – Erwan
    Commented May 12, 2022 at 18:10

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.