Training

A large component of the AAR-HPC project is the sharing of knowledge with the development and delivery of training.

MSU faculty and staff began developing a variety of online and in-person training courses for USDA-ARS scientists in January 2021. These training courses and workshops cover a range of geospatial, biologic, and data science topics and provide scientists an avenue to connect and build multi-disciplinary partnerships.

Current Training Courses and Workshops

Developing a UAS Program: This 4-hour workshop assists researchers in developing a robust UAS program to ensure their organization can take to the skies quickly, safely, efficiently, and cost effectively. The workshop will cover everything from planning a successful UAS research program and aircraft registration and certifications, to mission planning and data management. This workshop is ideal for those trying to decide if an unmanned aircraft system (UAS) is the appropriate equipment needed for their research.

UAS Missions, Operations, and Planning: This 4-hour workshop will provide research pilots of sUAS an overview of vehicle types, regulations, operations planning, flight skill and currency requirements, and reporting. Developing a safe and robust UAS program takes lots of planning and training, this workshop will help ensure your organization is prepared to take to the skies quickly, safely, and with a sound approach to gathering research data.

Data Wrangling: A Brief Primer on Data Collection, Processing, Storage, and Metadata: This 2-hour workshop will briefly touch on various common unmanned aerial system (UAS) payloads such as RGB and multispectral imagers, hyperspectral sensors, and LIDAR. It will then provide an overview of how photogrammetry works with RGB/multispectral imagery. A sample processing flow will be demonstrated in two of the popular commercial photogrammetry suites, Agisoft Metashape and Pix4D Pix4Dmapper, producing a reflectance-corrected and georeferenced orthomosaic from each software suite.

Introduction to Atlas: This 90-mintue information session is for anyone interested in learning more about USDA’s new state-of-the art supercomputer Atlas. Atlas is a high-performance computer with 101 terabytes of total RAM designed to help power research advances in biocomputing, epidemiology, geospatial technology and more. This introduction will cover a range of topics that include everything from a general overview of high-performance computing and why you might want to take advantage of its capabilities in your research to basic protocols associated with using Atlas (how to gain access to Atlas, how to log-in to your account, how to transfer data, etc.).

Blockchain Networks: This 4-hour workshop will provide a comprehensive introduction to blockchain networks (BCN). A BCN is a valuable platform for improving public confidence in data-driven decisions/policies. It can be viewed as a computer that is capable of executing algorithms that process data and output results. Topics covered will include a history of data processing, an introduction to useful BCN concepts (hash functions and digital signatures) in cryptography, which are essential for understanding how BCNs work, different types of BCNs (pros & cons), and implications on algorithms.

Introduction to Image Processing and Classical Machine Learning: This three-day workshop will cover the basics of image processing and classical machine learning using Python. Participants will be introduced to the basic concepts via PowerPoint lecture and guided through hands-on programming in an interactive Jupyter notebook framework. Participants will be encouraged to actively modify and expand the provided python code. Topics in image processing will include the basics of image processing, including conventions of image representation and image manipulations. Topics in classical machine learning will include the basics of feature extraction and labels; training, testing, and validation; and common methods for image classification. Topics in deep learning will include the basics of convolutional neural networks; training, testing, and validation in deep learning; and transfer learning.

Advanced Topics in Deep Learning: This two-day workshop will provide more in-depth exploration of some common deep learning architectures used in image processing. Participants will be introduced to the basic concepts via PowerPoint lecture and guided through hands-on programming in an interactive Jupyter notebook framework. Participants will be encouraged to actively modify and expand the provided python code. The first day will cover methods to explore, visualize, and modify network architectures. The second day will cover extensions to the convolutional neural network for such tasks as image segmentation, object detection, and spatio-temporal analysis.

Additional training courses will be developed throughout the duration of the project.