Skip to content

A tutorial session on pytorch lightning and tensorboard for better experimentation

Notifications You must be signed in to change notification settings

quentinf00/lightning-tensorboard-workshop

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pytorch Lightning & Tensorboard workshop

Slides

View Here

Edit Here

Exercise:

Open the exercise notebook in Colab and start the TODO items Open the notebook in colab

  • TODO 0 -> Log the metrics and hyperparams in tensorboard
  • TODO 1 -> Execute the training on GPU
  • TODO 2 -> Log the test confusion matrix image
  • TODO 3 -> save the best weight of the model
  • TODO 4 -> Log the graph of the model
  • TODO 5 -> Log the images of the worst prediction
  • TODO 6 -> Log the profile (GPU memory, CPU usage...) on a training step in tensorboard

Solutions:

The solutions can be found on the branches todo-[0-6]-sol And the final notebook on the branch all-solutions

Additional Material

If you're interested in knowing more, I highly encourage you to check out the pytorch lightning tutorials: https://pytorch-lightning.readthedocs.io/en/latest/#tutorials

About

A tutorial session on pytorch lightning and tensorboard for better experimentation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published