Research Computing Boot Camp & Vendor Presentations Spring 2022

January 18 – February 1, 2022

The Research Computing Services group welcomes you to attend their second Research Computing Boot Camp and associated vendor presentations!

The Research Computing Boot Camp gives you the opportunity to focus in more depth on R and MATLAB, both very useful for high performance computing.

In addition, Stata, ESRI (ArcGIS), NVIDIA, Wolfram (Mathematica), and MathWorks (MATLAB) advanced technical staff instructors will be offering exciting, highly useful sessions.

Registration notes:

1. There are a mix of in-person and over Zoom sessions. For the Zoom sessions, registrants for each session will be sent the Zoom link a few days before each session. In-person sessions will not be recorded; Zoom sessions will be recorded but these recordings may or may not be made available.
2. ArcGIS Online requires an account and we are creating accounts for those who need one. In addition, you will need access to the ArcGIS Pro desktop application. If you are attending the ArcGIS sessions and do not have either of these things, please fill out this form. Note that we need one week lead time to create accounts so please fill out the form early!
3. Register for as many sessions as you would like to attend.

Software installation notes:

• Many of the sessions are hands-on and, to get the most out of the tutorial, you should have the appropriate software installed on your computer before the tutorial starts.

This Boot Camp precedes our usual RCS Spring Tutorials, which will start on February 2; the schedule will be announced in early January. Our regular tutorials will not cover most of the topics covered here.

Boot Camp/Vendor Presentations Schedule

Boot Camp Topics

Building an R package (2-days workshop; Hands-on)

Instructor: Katia Bulekova (ktrn@bu.edu), BU Research Computing Services

Tuesday, January 18, 9:30 am – 12:30 am (with a 15 minute break) – 2 Cummington Mall, Room 107
Wednesday, January 19, 9:30 am – 12:30 am (with a 15 minute break) – 2 Cummington Mall, Room 107

These sessions have been cancelled and will be rescheduled.

Prerequisites: This workshop assumes that you are familiar with basic R and the RStudio IDE. This includes topics such as installing packages, assigning variables, and running basic R commands. You will need a recent version of R and RStudio. You can download R from https://cloud.r-project.org/ and RStudio from https://www.rstudio.com/products/rstudio/download/. You need to have an account on GitHub

Description: In this two-day workshop, you will learn the process for creating an R package from scratch. In addition, we will go over the best practices in building R packages, including how to test that functions execute properly, how to create documentation and examples, and how to add unit tests. Finally, we will cover using GitHub for publishing the new package. Please register for both days of this workshop.

Image Processing and Analysis: Unlocking the power of the Singular Value Decomposition (Hands-on)

Instructor: Josh Bevan (jbevan@bu.edu), BU Research Computing Services

Tuesday, January 18, 1:00 pm – 5:00 pm – 675 Commonwealth Avenue, SCI B39

This session has been cancelled and will be rescheduled.

Prerequisites: No prior MATLAB experience is assumed (although it will be helpful). You should be comfortable with calculus. Prior knowledge of Linear Algebra is helpful, but not required.

Description: In this Boot Camp we will explore the power of the Singular Value Decomposition (SVD) and how it is a very useful tool in Image Processing and Analysis. We will first explore the ideas underpinning the SVD and how it decomposes data into singular values and eigenvectors. We will then apply this knowledge to examine how it can be used for image compression, component analysis, facial recognition/classification with clustering, and feature detection. Throughout the Boot Camp attendees will work in small groups applying all these principles/techniques firsthand. In the last section of the Boot Camp each group will come up with a small project that applies the techniques learned in a novel way.

Natural Language Processing: With and without Neural Networks (Hands-on)

Instructor: Josh Bevan (jbevan@bu.edu), BU Research Computing Services

Wednesday, January 19, 10:00 am – 4:00 pm – 675 Commonwealth Avenue, SCI B39

This session has been cancelled and will be rescheduled.

Prerequisites: No prior MATLAB experience is assumed (although it will be helpful). No experience is required with Machine Learning, AI, Deep Learning etc.

Description: The vast majority of human communication and knowledge is encoded in “natural language”, this Boot Camp will focus on two techniques for “Natural Language Processing” (NLP) using computers. The first does not require explicit use of neural networks (NN), but the second does. We will first look at a way of encoding words and how this encoding carries with it semantic meaning, “word2vec”. We will explore how we can then use the encoded semantic meaning to turn mathematical operations into linguistic ones. Using simple addition and subtraction we will see how we can reconstruct analogies for example.

In the next part we will explore a breakthrough NN based NLP model called “GPT”. GPT first gained fame with the release of GPT-2, which is what we will explore using. GPT-2 simply tries to predict the next word in a sequence; however from this simple mechanism it is able to translate text, answer questions, summarizes passages, and generate text passages that can be indistinguishable from human-created ones. We will explore how to use GPT-2 based inference to do a variety of tasks.

Throughout the Boot Camp attendees will work in small groups applying all these principles/techniques firsthand. After the first and second section of the Boot Camp each group will come up with a small project that applies the techniques learned in a novel way.

Vendor Presentations

Introduction to Stata (Lecture)

Instructor: Chuck Huber, Director of Statistical Outreach, Stata

Monday, January 24, 11:00 am – 12:00 pm

In this talk I assume that the audience has never seen Stata before. I introduce the Stata interface, the data editor, the do-file editor, the menus, the dialog boxes, the help files, and the pdf documentation. I then briefly demonstrate how to use Stata to analyze data from cross-sectional, multilevel/longitudinal, and survival studies.

Introduction to Graphics with Stata (Lecture)

Instructor: Chuck Huber, Director of Statistical Outreach, Stata

Monday, January 24, 2:00 pm – 4:00 pm

This talk introduces the basics of using Stata graphics to explore your data, check the assumptions of your models, and present your results. I will demonstrate how to use -margins- and -marginsplot- to visualize the results of complex models. You will also learn how to customize the appearance of your graphs using graph schemes, formatting options, and how to layer and combine graphs.

It’s all Spatial!: Incorporating maps and spatial analysis into your humanities research (Hands-on)

Instructor: Brian Baldwin, Senior Solution Engineer – Education, ESRI

Tuesday, January 25, 10:00 am – 12:00 pm

This session has been switched from in-person on the Charles River Campus to over Zoom. The time has not changed. Registrants will be emailed the Zoom link.

Does the work or research in your field have anything to do with location? Of course it does! GIS has changed significantly in the last few years and it no longer requires a PhD in cartography to analyze data, create maps, and build immersive, interactive applications. This workshop will be a hands-on introduction to spatial tools and software that you already have access to at BU. It will also highlight recent updates to ArcGIS Online, ArcGIS StoryMaps, Esri field tools, and more. We will also be looking at examples from a wide range of disciplines (archeology, history, philosophy, and others). Most importantly, we will review how you can start leveraging these applications and tools in your discipline. This workshop is intended for a humanities audience.

ArcGIS Online requires an account and we are creating accounts for those who need one. In addition, you will need access to the ArcGIS Pro desktop application. If you do not have either of these things, please fill out this form. Note that we need one week lead time to create accounts so please fill out the form early!

Spatial Data Science: Bringing spatial tools and methodologies into your research (Hands-on)

Instructor: Brian Baldwin, Senior Solution Engineer – Education, ESRI

Tuesday, January 25, 1:00 pm – 3:00 pm

This session has been switched from in-person on the Charles River Campus to over Zoom. The time has not changed. Registrants will be emailed the Zoom link.

GIS has moved beyond plotting XY locations. Recent enhancements to Esri’s ArcGIS include deep learning tools for image analysis, the ability to use Jupyter Notebooks within ArcGIS Pro, the ability to connect to and incorporate open python libraries, among many others. This workshop will be a hands-on introduction to deep learning libraries and many other spatial tools and software that you already have access to at BU. Most importantly, we will review how you can start leveraging these applications and tools in your discipline. This workshop is intended for a physical and hard science audience.

ArcGIS Online requires an account and we are creating accounts for those who need one. In addition, you will need access to the ArcGIS Pro desktop application. If you do not have either of these things, please fill out this form. Note that we need one week lead time to create accounts so please fill out the form early!

NVIDIA: Using GPUs with Python (Hands-on)

Instructors: Brad Palmer, Senior Solutions Architect, Higher Education and Research, NVIDIA
Kristopher Keipert, Solutions Architect, NVIDIA

Wednesday, January 26, 2:00 pm – 4:00 pm

Python is the leading language for data science today and is increasingly being used in scientific computations. This workshop introduces Python GPU tools for porting and writing code that runs on GPUs. The primary tools, Numba and CuPy, are presented with coding examples. A Jupyter notebook is used for hands-on student exercises along with a set of lecture slides.

What’s New in Mathematica (Lecture)

Instructor: Kelvin Mischo, Certified Wolfram Instructor, Wolfram Research, Inc

Thursday, January 27, 10:00 am – 12:00 pm

With two or three major releases per year, Mathematica continues to grow in scope and now has built-in functions for deep learning, image and video processing, many aspects of data science, centralized repositories for user-defined functions and data sets, and greatly expanded options to share ideas or finished projects in web format, which are in addition to the ever-growing functionality related to mathematics. This seminar will show the scope of the built-in functions in Mathematica, and will focus on recently-released functionality. This seminar will show live calculations covering:

• Importing external data sets, and adding Wolfram curated data for additional insights
• Text scraping and text processing
• Creating web forms with live calculations that can be used by anyone (even people that don’t use Mathematica)
• Introduction to programming / loops, and tips for writing efficient code
• Machine learning and neural networks
• Additional examples across the language, especially newer functions, and useful documentation

No prior experience with Mathematica is required, and the content is suitable for faculty, staff, and students.

Data Science & Programming in Mathematica (Hands-on)

Instructor: Kelvin Mischo, Certified Wolfram Instructor, Wolfram Research, Inc

Thursday, January 27, 1:00 pm – 3:00 pm

Mathematica’s underlying language, the Wolfram Language, is a high-level language designed to give a user results very quickly with minimal coding or programming experience. This seminar will walk through several examples, and attendees will type along with the instructor to create a series of calculations in Mathematica. This seminar will show:

• A few basic calculations to show how Wolfram|Alpha technology is built into Mathematica
• Creating a mouse-driven interface
• Importing an external data file & creating visualizations
• Machine learning using that data set
• Examples of basic programming and loops, as well as parallel programming

Either Mathematica (local installation) or Mathematica Online are required, and can be requested here:
https://www.bu.edu/tech/services/cccs/desktop/distribution/mathsci/mathematica/

No prior experience with Mathematica is required, and the content is suitable for faculty, staff, and students.

Using Python with MATLAB (Lecture)

Instructor: Heather Gorr, Technical Marketing, Mathworks

Monday, January 31, 1:00 pm – 3:00 pm

This session has been switched from in-person on the Charles River Campus to over Zoom. The time has not changed. Registrants will be emailed the Zoom link.

Engineers who rely only on Python may find themselves encountering difficult or challenging tasks when it comes to embedded applications, building interactive dashboards, parallelizing applications, and deep learning. Contrarily, MATLAB is a full-stack advanced analytics platform that empowers domain experts to rapidly prototype ideas, validate models, and push applications into production with ease. However, sometimes it is advantageous to integrate MATLAB and Python together. One example being the need to combine MATLAB’s vast library of advanced analytics capabilities with supplemental models available in the open source community. Another, using Python as a language that is well suited to pipe data between different IT systems or the web.

There are several ways to integrate MATLAB and Python together either as R&D tools or as scalable components of your production infrastructure. The latter giving business users and decision makers immediate access to many of MATLAB’s built in analytics capabilities from deep learning, optimization, signal and image processing, computer vision, data mining, time-series forecasting, embedded code generation, and more.

In this session we demonstrate the many ways in which MATLAB and Python can interface and integrate with each other.

Highlights include:

• Calling Python libraries directly from MATLAB
• Calling Python from within a Simulink Model

Deep Learning with MATLAB (Hands-on)

Instructor: Ram Krishnamurthy, Senior Customer Success Engineer, Mathworks

Tuesday, February 1, 1:00 pm – 3:00 pm

This session has been switched from in-person on the Charles River Campus to over Zoom. The time has not changed. Registrants will be emailed the Zoom link.

Deep learning is quickly becoming embedded in everyday applications. It’s becoming essential for students and educators to adopt this technology to solve complex real-world problems. MATLAB and Simulink provide a flexible and powerful platform to develop and automate data analysis, deep learning, AI, and simulation workflows in a wide range of domains and industries:

In this hands-on workshop, you will write code and use MATLAB Online to:

1. Train deep neural networks on GPUs in the cloud
2. Create deep learning models from scratch for image and signal data
3. Explore pretrained models and use transfer learning
4. Learn how you can deploy your code to embedded targets
5. Discuss how you can interface with Python frameworks