Documentation
Search…
English
Weights & Biases
Quickstart
Examples
Guides
Reference
Command Line Interface
Python Library
Weave
App UI
Pages
Features
Panels
Custom Charts
Runs Table
Tags
Notes
Sweeps UI
Alerts
Teams
System Metrics
Anonymous Mode
Java Library
About Us
Homepage
Powered By
GitBook
System Metrics
Metrics automatically logged by wandb
wandb
automatically logs system metrics every 2 seconds, averaged over a 30 second period. The metrics include:
CPU Utilization
System Memory Utilization
Disk I/O Utilization
Network traffic (bytes sent and received)
GPU Utilization
GPU Temperature
GPU Time Spent Accessing Memory (as a percentage of the sample time)
GPU Memory Allocated
GPU metrics are collected on a per-device basis using
nvidia-ml-py3
. For more information on how to interpret these metrics and optimize your model's performance, see
this helpful blog post from Lambda Labs
.
Previous
Teams
Next
Anonymous Mode
Last modified
3mo ago
Copy link