![Szilard [Deeper than Deep Learning] on Twitter: "5/n My recommendations are still: If you don't have a GPU, lightgbm (CPU) is the fastest. If you have a GPU, xgboost (GPU) is also Szilard [Deeper than Deep Learning] on Twitter: "5/n My recommendations are still: If you don't have a GPU, lightgbm (CPU) is the fastest. If you have a GPU, xgboost (GPU) is also](https://pbs.twimg.com/media/D5fkhanUcAAD102.jpg:large)
Szilard [Deeper than Deep Learning] on Twitter: "5/n My recommendations are still: If you don't have a GPU, lightgbm (CPU) is the fastest. If you have a GPU, xgboost (GPU) is also
![Improving RAPIDS XGBoost performance and reducing costs with Amazon EMR running Amazon EC2 G4 instances | AWS Big Data Blog Improving RAPIDS XGBoost performance and reducing costs with Amazon EMR running Amazon EC2 G4 instances | AWS Big Data Blog](https://d2908q01vomqb2.cloudfront.net/b6692ea5df920cad691c20319a6fffd7a4a766b8/2020/07/17/EMRNVIDIA3.png)
Improving RAPIDS XGBoost performance and reducing costs with Amazon EMR running Amazon EC2 G4 instances | AWS Big Data Blog
![GPU-Accelerated Spark XGBoost - A Major Milestone on the Road to Large-Scale AI | NVIDIA Technical Blog GPU-Accelerated Spark XGBoost - A Major Milestone on the Road to Large-Scale AI | NVIDIA Technical Blog](https://developer-blogs.nvidia.com/wp-content/uploads/2019/10/Spark.png)