GC

Verified User in Consumer Goods

Mid-Market (51-1000 emp.)

7/26/2018

"XGB is the best out of the box model available today"

4.5/5

What do you like best about XGBoost?

Runs well in basically every situation, handles missing data well, relatively lightweight. Review collected by and hosted on G2.com.

What do you dislike about XGBoost?

It's annoying that you need graphviz to easily visualize the tree output. I can't believe it'd be that hard to output it even as a csv or something. Review collected by and hosted on G2.com.

TV

Tejasvini V.

Software Engineer

Computer Software

Small-Business (50 or fewer emp.)

10/25/2017

"Gradient Boosting Library For Multiple Languages"

5/5

What do you like best about XGBoost?

- It's portability as it supports Windows, Linux, OS X and also some cloud platforms.

- Multiple language support like: Python, R, Java and C++.

- Detailed documentation.

- There are various example in the documentation which really helps. Review collected by and hosted on G2.com.

What do you dislike about XGBoost?

To be frank, using this library with python till now didn't face any problems. Don't know for other languages but works well with python. Review collected by and hosted on G2.com.

Recommendations to others considering XGBoost:

If you are starting to learn it as well as a programming language too, start with python. It has great support for python and you will easily find tutorials. Review collected by and hosted on G2.com.

What problems is XGBoost solving and how is that benefiting you?

- Building Machine Learning based models.

- Developing projects and products that require Machine Learning algorithms as core. Review collected by and hosted on G2.com.

GH

Verified User in Higher Education

Small-Business (50 or fewer emp.)

11/10/2017

"A lot of documentation and great for ML competitions"

5/5

What do you like best about XGBoost?

The documentation makes it really easy to get started after the install. There are plenty of examples online to learn from and is widely regarded in the data community. I have only used xgboost for classification in a Kaggle competition and it made me interested in gradient boosted techniques. Review collected by and hosted on G2.com.

What do you dislike about XGBoost?

Directions for installing the package were not clear from what I remember. It can also be slow when training on a lot of data. LightGBM is another option for gradient boosting that trains much faster. Review collected by and hosted on G2.com.