Lightning-Universe/Training-Studio_app

Not able to run the sweeps

Madhu-gh opened this issue · 2 comments

Madhu-gh commented

🐛 Bug

I was not able to successfully run sweep using the example provided in Training Studio App

To Reproduce

  1. Go to Apps gallery
  2. Duplicate training studio App
  3. Run ---> lightning connect 01gkmref244yjm4zv3gqs860ef --yes
  4. Download script ---> curl https://raw.githubusercontent.com/Lightning-AI/lightning-hpo/master/sweep_examples/scripts/train.py -o train.py
  5. Run a sweep ---> lightning run sweep train.py --model.lr "[0.001, 0.01, 0.1]" --data.batch "[32, 64]" --algorithm="grid_search"

After few seconds, I get the following error:

image

Code sample

Expected behavior

Environment

  • PyTorch Version (e.g., 1.0):
  • OS (e.g., Linux):
  • How you installed PyTorch (conda, pip, source):
  • Build command you used (if compiling from source):
  • Python version:
  • CUDA/cuDNN version:
  • GPU models and configuration:
  • Any other relevant information:

Additional context

lantiga commented

Next time please copy-paste errors from the logs.

tchaton commented

Hey @Madhu-gh,

Thanks for reporting this bug.

After merging some fixes on lightning and making a release, everything should be working.

Can you confirm ?

Best,
T.C