Skip to content

Instantly share code, notes, and snippets.

@chychen

chychen/get_ready_modulus_bootcamp

Last active February 11, 2025 09:45
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save chychen/d19652961f3ba2977b0e28fbbe815aa4 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save chychen/d19652961f3ba2977b0e28fbbe815aa4 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
get_ready_modulus_bootcamp
setup bootcamp dev env, see https://gist.github.com/chychen/d19652961f3ba2977b0e28fbbe815aa4?permalink_comment_id=5433786#gistcomment-5433786
launch tensorboard on twcc, see https://gist.github.com/chychen/d19652961f3ba2977b0e28fbbe815aa4?permalink_comment_id=5433797#gistcomment-5433797
@chychen
Copy link
Author

chychen commented Feb 11, 2025 •
edited
Loading

Setup Bootcamp Env

Open up a terminal

image

Copy Paste following commands

# download bootcamp contents
git clone https://github.com/openhackathons-org/End-to-End-AI-for-Science
# lecture is designed based on /workspace path.
# so we remove it to make sure we could mount lecture materials on it.
# please make sure no important data within /workspace, otherwise please backup before removing it.
cd ~
sudo rm -rf /workspace 
sudo ln -s ~/End-to-End-AI-for-Science/workspace /.
cd /workspace/python
ls
# you should see 3 files listed: `Start_Here.ipynb  jupyter_notebook  source_code`

image

install relatives libs if missing

# Install libs
pip3 install gdown

go "End-to-End-AI-for-Science/workspace/python/Start_Here.ipynb to start

image

@chychen
Copy link
Author

chychen commented Feb 11, 2025 •
edited
Loading

Launch a Tensorboard

run tensorboard in a terminal

tensorboard --logdir=. --port 5000

image

setup TWCC port forwarding, click "Associate" button and check "5000" port

image
image

you will see a new associated port from target port 5000 to a new public port.

image

open up a new browser with public ip + public port (every user is different)

in above example, it is http://203.145.216.207:52482, then you should see the tensorboard is live.
image

check "Reload Data" to auto load logs after clicking on top-right "setting" button.

image

@chychen
Copy link
Author

chychen commented Feb 11, 2025

Open Modulus Container on TWCC

Go "Interactive Container"

image

Click "Create" button

image

find "Modulus"

image

Choose config with single GPU (default: 2 GPU), then click "Review & Create" -> "Create" -> done

image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment