Fall 2024 CRC Workshops
Fall 2024 Workshops
CRC Ecosystem On-ramp
Thursday Sept 12, 1-4 pm EST
This on-ramp session will introduce new and returning users to the CRC’s specialized compute resources and data storage systems. Topics covered include cluster access modalities, loading software tools with the LMOD module system, scheduling jobs with SLURM, and resource usage strategies.
Prerequisite: CRC user account
Instructors: Nick Comeau, Research Computing Specialist; Kim Wong, Research Associate Professor
Workshop Recording (hosted on Panopto)
Ab initio calculations for quantum-mechanical simulations
Thursday Sept 26, 1-4 pm EST
The worshop will be devoted to the use of quantum mechanical methods in the simulation of static, dynamical, and electronic properties of molecules and complex materials, with a focus on density-functional theory (DFT) based approaches. We will provide an introduction to the fundamental concepts of ab initio simulations, along with a discussion of different techniques and software available to solve specific problems, including the prediction of dynamical and vibration properties, ground- and excited-state electronic properties, and nuclear dynamics.
Prerequisite: A CRC user account will be helpful to test the examples provided during the workshop; a basic knowledge of quantum mechanics and solid state physics will be an advantage.
Instructors:Leonardo Bernasconi, Research Assistant Professor
Introduction to Programming for HPC with Python and Numba
Thursday Oct. 10, 1-4pm EST
This hands-on workshop offers an introduction to accelerating Python code using Numba, a powerful Just-In-Time (JIT) compiler. Learn how to achieve high efficiency by exploiting powerful GPU and CPU compute capabilities with minimal changes to your existing Python code, enabling you to harness the full potential of your hardware. Through practical examples, discover how Numba simplifies the development process, making it easier to optimize performance-critical sections of your applications. Whether you're new to HPC or looking to enhance your Python skills, this workshop provides the tools you need to write fast, efficient code with ease.
Prerequisites: Intermediate level Python
Instructor: Chengnian Xiao, Engineering HPC Consultant
Link to workshop recordings
Link to latest slides
LLMs Customization and Deployment with RAG and LoRA
Thursday Oct. 24, 1-4pm EST
This workshop will explore various LLM customization pipelines, including the development of retrieval-augmented generation (RAG) agents for LLMs, which involve using vector stores and similarity search. It will also cover parameter-efficient fine-tuning techniques, such as Low-Rank Adaptation (LoRA). Additionally, the workshop will focus on building inference servers with these customized models and deploying them on on-premise hardware using Docker and the NVIDIA Container Toolkit.
Prerequisite: Introductory deep learning knowledge, with comfort with PyTorch and transfer learning preferred. Intermediate Python experience, including object-oriented programming and libraries. Familiarity with containerization tools such as Docker.
Instructor: Yassin Khalifa, Data Scientist
Workshop Slides
Workshop Materials
Workshop Recording
MONAI
Thursday, Nov. 14, 1-4pm EST
This workshop will explore Project MONAI, a set of open-source, freely available collaborative frameworks built for accelerating research and clinical collaboration in Medical Imaging. The goal of MONAI is to accelerate the pace of innovation and clinical translation by building a robust software framework that benefits nearly every level of medical imaging, deep learning research, and deployment.
Prerequisite: Introductory deep learning knowledge, with comfort with PyTorch and transfer learning preferred. Intermediate Python experience, including object-oriented programming and libraries.
Instructor: Yassin Khalifa, Data Scientist