argonne-lcf/ATPESC_MachineLearning

Lecture and hands-on material for Track 6 - AI/ML - of Argonne Training Program on Extreme-Scale Computing

Python

433 commits

updated Aug 7, 2026

extremecomputingtraining.anl.gov

See the code

README

ATPESC 2026

Agenda

Monday, August 3

Time	Talk	Speaker
8:30AM	Welcome and Introduction	Shilpika, ANL
8:35AM	Transition time: splitting into groups (people new to deep learning vs. more experienced)
8:35AM	Parallel Session
	- Main room: Introduction to Deep Learning	Huihuo Zheng, ANL
	- Breakout room: Distributed Deep Learning	Bethany Lusch, ANL
9:50AM	Break
10:15AM	Profiling Deep Learning	Nathan Nichols, ANL
11:20 AM	Introduction to Large Language Models (LLMs)	Jingyan (Jane) Jiang, ANL
12:00PM	Pre-training on a Supercomputer - Part 1	Sam Foreman, ANL
12:30PM	Lunch
1:30PM	Pre-training on a Supercomputer - Part 2	Sam Foreman, ANL
2:30PM	Post-training (Finetuning/Alignment/RL techniques)	Filippo Simini, ANL
4:00PM	Explainable AI for Science	Shilpika, ANL
4:30PM	Break
5:30PM	Featured Speaker	Jack Dongarra

Tuesday, August 4

Time	Talk	Speaker
8:30AM	Welcome and introduction	Shilpika, ANL
8:35AM	Workflow Tools for Science	Christine Simpson, ANL
9:30AM	Coupled Workflows for Science (Simulations + AI Workflows)	Riccardo Balin and Christine Simpson, ANL
10:30AM	Break
10:50AM	Inference	Misha Salim, ANL
11:50 AM	Agentic Tools - Part 1 (Trinity/Hermes/OpenClaw)	Huihuo Zheng, ANL
12:30 PM	Lunch
1:30 PM	AI Testbed	Varuni Sastry and Murali Emani, ANL
3:00PM	Agentic Workflows for Science	Thang Pham, ANL
4:00PM	Agentic Tools - Part 2 (Academy)	Kyle Chard, UChicago/ANL
4:30PM	Break
5:30PM	Featured Speaker	Bill Gropp

At the beginning of the first day, we will temporarily split into two groups. Attendees can choose between Introduction to Deep Learning and Distributed Deep Learning.

The "Introduction to deep learning" session will rely on Jupyter Notebooks which are targeted for running on Google's Colaboratory Platform or ALCF JupyterHub. The Colab platform gives the user a virtual machine in which to run Python codes including machine learning codes. The VM comes with a preinstalled environment that includes most of what is needed for these tutorials.

The other sessions involve Python scripts executed on the Aurora and AI Testbed platforms at ALCF.

Reservations

Queue:
- Daytime reservations: -q ATPESC
- Evening reservations: -q ATPESC-Night
- Outside of reservations: -q debug or -q prod (more info)
Project/Allocation: ATPESC2026 (-A ATPESC2026)
Shared directories:
- Aurora: /flare/ATPESC2026
- Polaris: /eagle/ATPESC2026
ALCF Unix Groups: ATPESC2026

Using Google Colab

Google Colab involves running Jupyter notebooks.

Do the following before you come to the tutorial:

You need a Google Account to use Colaboratory
Go to Google's Colaboratory Platform
You should see this page
Now you can open the File menu at the top left and select Open Notebook which will open a dialogue box.
Select the GitHub tab in the dialogue box.
From here you can enter the url for the github repo: https://github.com/argonne-lcf/ATPESC_MachineLearning and hit <enter>.
This will show you a list of the Notebooks available in the repo. When you select a notebook from this list it will create a copy for you in your Colaboratory account (all *.ipynb files in the Colaboratory account will be stored in your Google Drive).
To use a GPU in the notbook select Runtime -> Change Runtime Type and select an accelerator.

Weights & Biases API key

For the Pre-training on a Supercomputer session, you will need a Wandb api_key. Visit https://docs.wandb.ai/quickstart/ to sign-up and get the key.

Similar repos

henryhxu/CSCI5540

Grad course on ML systems

wandb/edu

Educational materials on deep learning by Weights & Biases

693

DrMuratAltun/llm-bootcamp

LLM Tabanlı Uygulama Geliştirme Atölyesi · BTK Akademi · Dr. Murat Altun · 42 saat ileri seviye…

A-Alwabel/AI-Diploma-Program

Comprehensive AI Diploma Program - Course Materials (Solutions Excluded)

little51/llm-workshop

大模型应用与开发实验教程

jdmartinev/ArtificialIntelligenceIM

NAU-CS/ml-group-meetings

mryab/efficient-dl-systems

Efficient Deep Learning Systems course materials

1,038

Languages

Python

55.0%

Jupyter Notebook

24.8%

HTML

16.1%

Shell

4.1%

argonne-lcf/ATPESC_MachineLearning

Lecture and hands-on material for Track 6 - AI/ML - of Argonne Training Program on Extreme-Scale Computing

Python

433 commits

updated Aug 7, 2026

extremecomputingtraining.anl.gov

See the code

README

ATPESC 2026

Agenda

Monday, August 3

Time	Talk	Speaker
8:30AM	Welcome and Introduction	Shilpika, ANL
8:35AM	Transition time: splitting into groups (people new to deep learning vs. more experienced)
8:35AM	Parallel Session
	- Main room: Introduction to Deep Learning	Huihuo Zheng, ANL
	- Breakout room: Distributed Deep Learning	Bethany Lusch, ANL
9:50AM	Break
10:15AM	Profiling Deep Learning	Nathan Nichols, ANL
11:20 AM	Introduction to Large Language Models (LLMs)	Jingyan (Jane) Jiang, ANL
12:00PM	Pre-training on a Supercomputer - Part 1	Sam Foreman, ANL
12:30PM	Lunch
1:30PM	Pre-training on a Supercomputer - Part 2	Sam Foreman, ANL
2:30PM	Post-training (Finetuning/Alignment/RL techniques)	Filippo Simini, ANL
4:00PM	Explainable AI for Science	Shilpika, ANL
4:30PM	Break
5:30PM	Featured Speaker	Jack Dongarra

Tuesday, August 4

Time	Talk	Speaker
8:30AM	Welcome and introduction	Shilpika, ANL
8:35AM	Workflow Tools for Science	Christine Simpson, ANL
9:30AM	Coupled Workflows for Science (Simulations + AI Workflows)	Riccardo Balin and Christine Simpson, ANL
10:30AM	Break
10:50AM	Inference	Misha Salim, ANL
11:50 AM	Agentic Tools - Part 1 (Trinity/Hermes/OpenClaw)	Huihuo Zheng, ANL
12:30 PM	Lunch
1:30 PM	AI Testbed	Varuni Sastry and Murali Emani, ANL
3:00PM	Agentic Workflows for Science	Thang Pham, ANL
4:00PM	Agentic Tools - Part 2 (Academy)	Kyle Chard, UChicago/ANL
4:30PM	Break
5:30PM	Featured Speaker	Bill Gropp

At the beginning of the first day, we will temporarily split into two groups. Attendees can choose between Introduction to Deep Learning and Distributed Deep Learning.

The other sessions involve Python scripts executed on the Aurora and AI Testbed platforms at ALCF.

Reservations

Queue:
- Daytime reservations: -q ATPESC
- Evening reservations: -q ATPESC-Night
- Outside of reservations: -q debug or -q prod (more info)
Project/Allocation: ATPESC2026 (-A ATPESC2026)
Shared directories:
- Aurora: /flare/ATPESC2026
- Polaris: /eagle/ATPESC2026
ALCF Unix Groups: ATPESC2026

Using Google Colab

Google Colab involves running Jupyter notebooks.

Do the following before you come to the tutorial:

You need a Google Account to use Colaboratory
Go to Google's Colaboratory Platform
You should see this page
Now you can open the File menu at the top left and select Open Notebook which will open a dialogue box.
Select the GitHub tab in the dialogue box.
From here you can enter the url for the github repo: https://github.com/argonne-lcf/ATPESC_MachineLearning and hit <enter>.
This will show you a list of the Notebooks available in the repo. When you select a notebook from this list it will create a copy for you in your Colaboratory account (all *.ipynb files in the Colaboratory account will be stored in your Google Drive).
To use a GPU in the notbook select Runtime -> Change Runtime Type and select an accelerator.

Weights & Biases API key

For the Pre-training on a Supercomputer session, you will need a Wandb api_key. Visit https://docs.wandb.ai/quickstart/ to sign-up and get the key.

Similar repos

henryhxu/CSCI5540

Grad course on ML systems

wandb/edu

Educational materials on deep learning by Weights & Biases

693

DrMuratAltun/llm-bootcamp

LLM Tabanlı Uygulama Geliştirme Atölyesi · BTK Akademi · Dr. Murat Altun · 42 saat ileri seviye…

A-Alwabel/AI-Diploma-Program

Comprehensive AI Diploma Program - Course Materials (Solutions Excluded)

little51/llm-workshop

大模型应用与开发实验教程

jdmartinev/ArtificialIntelligenceIM

NAU-CS/ml-group-meetings

mryab/efficient-dl-systems

Efficient Deep Learning Systems course materials

1,038

Languages

Python

55.0%

Jupyter Notebook

24.8%

HTML

16.1%

Shell

4.1%