XGBoost Profiling and Monitoring
See the Quick Start Guide on how to install and configure the profiler.
To profile XGBoost training, use a callback. Only some of the iterations will be profiled; the profiler decides which iterations to profile for optimal statistics and low overhead.
from graphsignal.profilers.xgboost import GraphsignalCallback bst = xgb.train(..., callbacks=GraphsignalCallback()])
The XGBoost Iris example illustrates how to add the profiler callback.
Graphsignal provides built-in support for distributed training and inference. Depending on the platform, it may be necessary to provide a run ID to all workers. Refer to Distributed Workloads section for more information.
After starting the run, profiling and monitoring data will be available in the cloud dashboards.