XProf Demo

Open in Colab

This is a simple demonstration of XProf's capabilites, on an example training workload running on Cloud TPUs.

# Install the stable version of the profiler plugin
pip install -U tensorboard_plugin_profile
 
# git clone the xprof repo so we have access to the demo data there
git clone http://github.com/openxla/xprof
 
Cloning into 'xprof'...
warning: redirecting to https://github.com/openxla/xprof/
remote: Enumerating objects: 12360, done.
remote: Counting objects: 100% (428/428), done.
remote: Compressing objects: 100% (257/257), done.
remote: Total 12360 (delta 244), reused 214 (delta 170), pack-reused 11932 (from 2)
Receiving objects: 100% (12360/12360), 52.22 MiB | 36.28 MiB/s, done.
Resolving deltas: 100% (9127/9127), done.

# Load the TensorBoard notebook extension.
%load_ext tensorboard

# Launch TensorBoard and navigate to the Profile tab to view performance profile
%tensorboard --logdir=xprof/demo

Once tensorboard loads the profile data, use the tools drop down to explore various tools. Please see the per-tool documentation pages for an explanation of what you're seeing.