Difference between revisions of "ALCF ML Workshop"
From SCECpedia
Jump to navigationJump to search| (2 intermediate revisions by the same user not shown) | |||
| Line 5: | Line 5: | ||
*[https://www.alcf.anl.gov/alcf-resources/theta ALCF Theta Description]] | *[https://www.alcf.anl.gov/alcf-resources/theta ALCF Theta Description]] | ||
*[https://github.com/argonne-lcf/sdl_ai_workshop Workshop Github Repo] | *[https://github.com/argonne-lcf/sdl_ai_workshop Workshop Github Repo] | ||
| + | *[https://github.com/jonasrauber/eagerpy Useful ML Github repo] | ||
| + | *[https://anl.app.box.com/s/zqo4v52e4bnl3ox8bvqtajjiuro2uha2 Getting Started at ALCF] | ||
== Use of Google Colaboratory == | == Use of Google Colaboratory == | ||
| Line 10: | Line 12: | ||
*[https://colab.research.google.com/notebooks/intro.ipynb#recent=true Test Python Jupyter Notebook] | *[https://colab.research.google.com/notebooks/intro.ipynb#recent=true Test Python Jupyter Notebook] | ||
*<username>@theta.alcf.anl.gov | *<username>@theta.alcf.anl.gov | ||
| + | |||
| + | == qsub == | ||
| + | From /home/maechlin/sdl_ai_workshop/01_distributedDeepLearning/Horovo | ||
| + | *qsub -O pytorch_keras_scale -n 128 -q training -A SDL_Worsubmissions/theta/qsub_keras_mnist_scale.sh | ||
| + | |||
| + | From /home/maechlin/sdl_ai_workshop/02_dataPipelines/00_tensorflowDatasetA | ||
| + | *qsub -O tensorflow_dataset -n 1 -q training -A SDL_Workshop submit_theta.sh | ||
== SCEC At ALCF == | == SCEC At ALCF == | ||
Latest revision as of 20:16, 8 December 2020
SCEC researchers participated in the ALCF ML Workshop 8 Dec - 10 Dec 2020
ALCF ML Workshop
Use of Google Colaboratory
Workshop will use jupyter notebooks running in google collaboratory.
- Test Python Jupyter Notebook
- <username>@theta.alcf.anl.gov
qsub
From /home/maechlin/sdl_ai_workshop/01_distributedDeepLearning/Horovo
- qsub -O pytorch_keras_scale -n 128 -q training -A SDL_Worsubmissions/theta/qsub_keras_mnist_scale.sh
From /home/maechlin/sdl_ai_workshop/02_dataPipelines/00_tensorflowDatasetA
- qsub -O tensorflow_dataset -n 1 -q training -A SDL_Workshop submit_theta.sh
