Hoffman2:JupyterNotebook: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
(→Method 1:: fix broken links, restructure content, and add GPU information) |
||
| Line 1: | Line 1: | ||
=Use Jupyter Notebook/Lab= | |||
==Install on Local Device== | |||
Install to your laptop using Anaconda: | |||
[https://jupyter.readthedocs.io/en/latest/install.html Jupyter Official Website] | |||
#Download [https://www.anaconda.com/distribution/ Anaconda] version 3.7 | |||
#Install it to your laptop as a regular application | |||
#then run "jupyter notebook" from the terminal | |||
==Connecting to Hoffman2== | |||
Download h2jupynb from [https://gitlab.idre.ucla.edu/dauria/jupyter-notebook git ] to your local drive | Download h2jupynb from [https://gitlab.idre.ucla.edu/dauria/jupyter-notebook git ] to your local drive | ||
| Line 12: | Line 22: | ||
./h2jupynb -u userid -v 3.7.0 | ./h2jupynb -u userid -v 3.7.0 | ||
(You'll find more details [https://www.hoffman2.idre.ucla.edu/ | (You'll find more details [https://www.hoffman2.idre.ucla.edu/Using-H2/Connecting/Connecting.html#connecting-via-jupyter-notebook-lab here]) | ||
===GPU Access=== | |||
To access GPU resources on Hoffman, add the virtual env to ipykernel for the very first time: | |||
qrsh -l gpu,P4 | |||
module load python/anaconda3 | |||
conda activate pytorch-1.3.1-gpu | |||
python -m ipykernel install --user --name=pytorch-1.3.1-gpu | |||
Using h2jupynb: | |||
./h2jupynb -v anaconda3 -g yes -c P4 -l 10.0 | |||
Revision as of 07:21, 26 October 2020
Use Jupyter Notebook/Lab
Install on Local Device
Install to your laptop using Anaconda:
- Download Anaconda version 3.7
- Install it to your laptop as a regular application
- then run "jupyter notebook" from the terminal
Connecting to Hoffman2
Download h2jupynb from git to your local drive
chmod u+x h2jupynb
From your laptop, run (Using Python 3.7)
./h2jupynb -u userid -v 3.7.0
(You'll find more details here)
GPU Access
To access GPU resources on Hoffman, add the virtual env to ipykernel for the very first time:
qrsh -l gpu,P4 module load python/anaconda3 conda activate pytorch-1.3.1-gpu python -m ipykernel install --user --name=pytorch-1.3.1-gpu
Using h2jupynb:
./h2jupynb -v anaconda3 -g yes -c P4 -l 10.0