This workshop will demonstrate how to use Matlab to create data visualization and animation for publications or presentations.

The talk is the slide presentation in UCLA OARC/IDRE workshop "Learning PyTorch" in Fall 2021.

Slides and demos can be downloaded from the github repo: https://github.com/huqy/deep_learning_workshops

VIDEO CHAPTERS
00:00 Opening Remarks
01:14 What is PyTorch
06:47 Why PyTorch
14:40 PyTorch Tensors
22:02 Py and Non-Py in PyTorch
28:21 Automatic differentiation
33:06 Optimizers in PyTorch
36:34 Neural networks in PyTorch
44:11 PyTorch, Tensorflow and JAX

This workshop will demonstrate how to use Matlab to create data visualization and animation for publications or presentations.

This is the slide presentation for UCLA OARC/IDRE Workshop "Learning Scikit-Learn: Basics" in Summer 2021.

The pdf slide and the demo files can be downloed from the github repo: https://github.com/huqy/learning-sklearn

Video Chapters:
00:00 Opening remarks
03:25 Knowing scikit-learn in minutes
12:21 Simplified workflow for a machine learning project
15:09 Data input, loader and generator
20:20 Splitting the data
24:51 Scikit-learn core objects
29:11 Data pre-processing
32:08 Models and meta-estimators
43:35 Model persistence
48:29 Scikit-learn extension libraries
53:32 Key Takeaways

IDRE Workshop from August 6, 2021
Materials available here: https://github.com/benjum/idre-intro-to-shell-scripting

IDRE Workshop from July 30, 2021
Materials available here: https://github.com/benjum/idre-intro-to-shell-cli

IDRE Workshop from January 15, 2021
Materials available here: https://github.com/benjum/numerical-python-part2

IDRE Workshop from January 8, 2021
Materials available here: https://github.com/benjum/numerical-python-part1

IDRE Workshop from January 22, 2021
Materials available here: https://github.com/benjum/numerical-python-part3

Video contents:
0:00:00 Introduction
0:18:46 Secure shell (ssh)
0:26:28 ssh key pair: demo
0:34:05 Linux Shell environment
0:46:49 Vim text editor
1:00:54 Vim: basic editing
1:09:10 Vim: browsing large code base
1:15:05 Git version control
1:23:04 Git merge conflict resolution using Vim
1:29:39 Git rebase using Vim: explanation
1:36:06 Git rebase using Vim: demo
1:39:02 Tmux: terminal multiplexer

Supplementary course materials:
https://github.com/schuang/essential-command-line-tools-tutorial

Questions and comments:
https://github.com/schuang/essential-command-line-tools-tutorial/issues

* We are in the process of fixing the wrong words in the subtitles. Thank you for your understanding.

Instructor: Yoh Kawano

Workshop materials: https://github.com/benjum/idre-data-sci-with-python

Part 2: A Data Science case study: 3D Mapping of Los Angeles arrest data

- libraries to include: sodapy, pandas, plotly, keplergl

Machine learning with iris dataset

- examples with pytorch and tensorflow

Workshop material: https://github.com/benjum/idre-data-sci-with-python

Part 1: Introducing data science with python using the iris dataset to cover topics for getting, cleaning, and visualizing data.

libraries to include: openpyxl, csv, json, requests, urllib, pandas, matplotlib, seaborn, plotnine, plotly, altair

IDRE Workshop from July 28, 2020
Materials available here: https://github.com/benjum/idre-intro-to-sql

Cloud computing service is the delivery of computing resources including applications, servers, storage, databases, networking, analytics, and artificial intelligence over the internet. This workshop series is intended to provide hands-on tutorials for academic researchers to use cloud computing resources from two major cloud providers, Amazon Web Service and Google Cloud Platform.

Cloud computing service is the delivery of computing resources including applications, servers, storage, databases, networking, analytics, and artificial intelligence over the internet. This workshop series is intended to provide hands-on tutorials for academic researchers to use cloud computing resources from two major cloud providers, Amazon Web Service and Google Cloud Platform.

Cloud computing service is the delivery of computing resources including applications, servers, storage, databases, networking, analytics, and artificial intelligence over the internet. This workshop series is intended to provide hands-on tutorials for academic researchers to use cloud computing resources from two major cloud providers, Amazon Web Service and Google Cloud Platform.

Anaconda3 is comprehensive and versatile solution for scientific computing and data analysis. It provides a free and open-source platform that supports many packages written in different languages. In this workshop, we will explore how to use Anaconda3 to install and manage these packages efficiently and effectively.

Speaker: Seyoon Ko, Ph.D.
IDRE Fellow and Assistant Adjunct Professor,
Mathematics Department,
University of California Los Angeles

Abstract: Julia (http://julialang.org) is a modern open-source programming language for technical computing. Its design offers much greater speed and productivity compared to R or Python, as high-performance code does not need to be wrapped in a low-level language like C or Fortran. After almost a decade of active development, Julia reached its first major release, v1.0, in 2018, and is quickly gaining popularity in scientific computing and data science communities. In this workshop, I will present the basic concepts of Julia and show a little comparison between Julia and other languages, such as R, C, and Python.

The IDRE Early Career Researchers group's panel is a discussion on how high-performance computing (HPC) is used across STEM disciplines. It is led by the following new cohort of IDRE fellows (2023-2024) from various disciplines across the UCLA campus.

Robert Fofrich – Institute of the Environment and Sustainability
Fei-man Hsu – Department of Molecular, Cell, and Development Biology
Alec Linot – Mechanical and Aerospace Engineering
Xiao Luo – Department of Computer Science
Namjoon Suh – Department of Statistics and Data Science

In this presentation, we will explore how the HPC can be used to accomplish common DH tasks, such as network and GIS analysis with large data sets which would be computationally difficult or expensive on typical desktop / laptop hardware. We will use least-cost pathing with GDAL, network analysis with Python, and general DH data analysis and visualization with Jupyter Notebooks. This workshop is designed to highlight the usability and accessibility of these common tools in the HPC environment.

Machine learning has undeniably emerged as a crucial element in the field of science and technology in recent years. Consequently, mastering libraries like scikit-learn, pivotal in the arsenal of machine learning tools, is of paramount importance.In this session, we delve into advanced aspects of leveraging the scikit-learn Python library for high-performance machine learning. We will focus on enhancing modeling speed with multicore processors and accelerators, as well as implementing out-of-core learning. This lecture builds upon the foundational concepts covered in the first session. For optimal benefit, attendees should possess practical experience in Python programming and a solid understanding of fundamental machine learning concepts.

https://idre.ucla.edu/calendar-event/high-performance-machine-learning-using-scikit-learn-2

Slides and Demos: https://github.com/huqy/learning-sklearn

By Benjamin Winjum, Project Scientist, OARC
December 7, 2023, 10-11AM via Zoom

The application of machine learning (ML) and high performance computing (HPC) to humanities research holds tremendous potential, but there are significant challenges to developing a community of experts that can work together at this intersection of topic and technique. HPC experts may not have the background necessary to appreciate how their tools can best be brought to bear on the diverse datasets and conceptual approaches of humanities researchers, and humanities researchers may not have the well-honed background in computational and statistical modeling that resides with ML or HPC specialists. We will approach this topic by focusing on a few case studies in the humanities that illustrate how ML and HPC can both be used to powerful effect, whether by streamlining the initial processing of massive data, modeling and visualizing humanities-relevant datasets, or using deep learning frameworks and the speed and power of GPUs to rapidly prototype predictive models.

Abstract: Human voice production is a complex process requiring precise coordination of respiratory and laryngeal muscle activation. These components shape voice quality, enabling humans to produce a wide range of voice types. However, their interaction remains poorly understood, mainly due to the challenges in directly observing the larynx. The goal of our study is to identify physiological factors contributing to individual voice differences. We propose a computational model to simulate human laryngeal muscle activation in the phonation process. By modeling laryngeal muscle activation in male and female larynges, we examine how different muscles coordinate to posture the vocal folds in producing voices of diverse voice qualities.

Speaker: Tsukasa Yoshinaga, Ph.D.
IDRE Postdoctoral Fellow
Department of Head and Neck Surgery
University of California Los Angeles

Thursday, October 14, 2021 from 12-2pm

Scholarship of visual materials requires close looking and comparison across works. Digitization efforts across the world have allowed for many digital facsimiles of these materials to be made available online. For scholarship, digital viewers are extremely valuable to presenting high-resolution imaging, which allows for close examination of materials. Interoperability has become crucial for institutions for contrasting objects from different collections. Increasing stability in digital visual scholarship means working from authority files from museums, libraries, and archives — institutions are committed to long term preservation and standards.

This workshop will introduce participants to Mirador, an “open-source, web based, multi-window image viewing platform with the ability to zoom, display, compare and annotate images from around the world.” Leveraging the International Image Interoperability Framework (IIIF), the Mirador web-viewer allows users to compare and annotate images and documents. Participants will learn how to install a local instance of Mirador, find Mirador-friendly content and resources, make narratives with annotations, and complete side-by-side transcriptions.

All levels are welcome.

Augmented reality (AR) is the layering of computer-generated perceptual information onto real environments. The experience requires a level of interactivity and sensory engagement. AR is being adopted across the fields of architecture and engineering, as well as adopted by artists to make creative connections between our shared reality and that of innumerable virtual worlds.

This workshop will introduce participants to the design and mechanics of augmented reality. We will focus on working with images, bringing digital objects into live environments through the use of the ArtVive app. Attendees will create free Artivive accounts and learn to create an AR animation. No prior technical experience necessary.

IDRE Workshop from April 24, 2021
Materials available here: https://github.com/benjum/idre-spring21-python-data-viz-2

IDRE Workshop from April 16, 2021
Materials available here: https://github.com/benjum/idre-spring21-python-data-viz-1

3D models can be formed in a number of ways. Two common surveying methods that result in 3D models are photogrammetry and light detection and ranging or lidar for short.

A form of photogrammetry has been around since the beginnings of photography in the mid-19th century. The term usually refers to the method by which 2D images are compared and combined using optics, projective geometry, and computing to produce a 3D model. This method of image overlap to create a dimensional model is applied in many object-oriented disciplines such as archeology, topographic mapping, architecture, engineering, manufacturing, quality control, police investigation, cultural heritage, and geology.

Lidar is a method for measuring distances (ranging) by illuminating the target with laser light and measuring the reflection with a sensor. Differences in laser return times and wavelengths can then be used to make digital 3D representations of the target. While long used in the earth sciences, lidar is becoming increasingly available to all as these sensors are being added to cellular mobile devices, increasing its academic and commercial potential.

In this workshop, participants will learn about the hardware and software required for these methods. We will also cover the workflow that is involved in producing photogrammetric and lidar 3D models. All levels welcome.

If you have any further questions regarding the workshop, please contact the instructor, Dr. Francesca Albrezzi at falbrezzi@ucla.edu

While we are working remotely, there are many ways that we can share virtual space. Mozilla Hubs is a Web-based platform that provides users with three dimensional virtual space that can allow for links, videos, 2D and 3D objects, and live interaction. As an alternative to zoom, Mozilla Hubs has been used for poster sessions, conferences, performances, exhibitions, and more. It can also be accessed through a virtual reality headset for an immersive experience.

In this workshop, participants will be introduced to Mozilla Hubs and how to use the built-in functionality. For more customized environments, we will also cover how to build your own basic environment and bring it into Mozilla Hubs through Spoke, Mozilla Hub’s web editor.



If you have any further questions regarding the workshop, please contact the instructor, Dr. Francesca Albrezzi at falbrezzi@ucla.edu.

https://idre.ucla.edu/calendar-event/creating-vr-environments-mozilla-hubs-and-blender-basics

August 11, 2020 @ 2:00 pm - 4:00 pm

We are increasingly sharing our data and research in virtual environments. WebVR (or WebXR as it’s now known) is an open standard that allows 3D / virtual reality (VR) experiences to be produced through your browser. It also aims to make your immersive data platform-agnostic, so it will be portable to whatever device you have. In this workshop, you will be introduced to ways to graph three dimensional data. You will also be shown how to bring that 3D graph into a browser environment so that your work can be shared through the Web.In the hands-on segment of the workshop, we will be using the following tools:
R
A-Frame
Jupyter Notebooks
There are no pre-requisite requirements to take this workshop.


If you have any further questions regarding the workshop, please contact instructor Francesca Albrezzi.

Workshop by Francesca Albrezzi, PhD
Credit to Miriam Posner, PhD for several slides from her DH101 lectures

Workshop slides: http://bit.ly/DataVizSlidesF2020

TOC
3:56: Overview on Data Types and Visualization Methods
31:33 Open Refine Reconciliation Using the Getty Vocabularies
1:08:33 Google Pivot Tables
1:17:30 Tableau

The digital realm is becoming an increasingly popular space for creatives. By creatives, I don't just mean artists, but anyone who wants to harness their creativity, skill, and imagination to produce something with a personal touch and flair. From interactivity to detailed design, browser windows can connect with viewers in powerful ways. So how do you start making the computer screen your canvas? It starts with programming.

Programming is writing instructions that computers can understand and execute. For more than fifty years, computer programmers have been writing code and there are now more than 2,500 documented programming languages. So which language do you start with? What are the differences?

In this workshop, you will receive an introduction to the basics of programming. Specifically, we will cover:

-What is programming
-What is processing
-The difference between Low-Level vs High Level programming languages
-The difference between Client Side vs. Service Side Programs
-p5.js, a free and open-source JavaScript library for creative coding, with a focus on making coding accessible and inclusive for artists, designers, educators, and beginners
-How to use the p5.js web editor
-How to create shapes and draw using p5.js

All levels welcome and no experience required.

If you have any further questions regarding the workshop, please contact the instructor, Dr. Francesca Albrezzi.

QGIS workshop for processing and visualizing demographic census data, commonly used in the social sciences.

An Introduction to QGIS and Digital Gazetteers

Instructor: Yoh Kawano
Course material:
https://github.com/yohman/workshop-leaflet

Course setup:
https://github.com/yohman/workshop-leaflet/blob/main/setup.md

Course lab:
https://github.com/yohman/workshop-leaflet/blob/main/lab.md

Research projects are often confronted with the need to publish results to communicate their findings to the general audience. Publishing research on an interactive web-based map is an effective and visually compelling way to do so.

In this workshop, participants will learn the basics of HTML, CSS and Javascript. We will use the Leaflet library to overlay research data and create compelling narrative maps. Finally, participants will learn how to use GitHub Pages as a publishing platform for the interactive maps.

Yoh Kawano Ph.D., Lead Computational Scientist, GIS/Spatial Data Science - UCLA Office of Advanced Research Computing

Geospatial platforms have dramatically evolved over the past two decades. What's the human story and impact behind them? This talk will highlight the history of innovative and influential projects designed through academic, government, and community partnerships at UCLA. It will discuss the impact they have had upon our communities and how they address the digital divide (e.g. Neighborhood Knowledge Los Angeles, HyperCities, Million Dollar Hoods). The talk will also highlight a personal journey into spatial ethnography, and how in-depth, on-the-ground fieldwork can address healing in a risk society and illuminate the narratives of people affected by catastrophic disaster situations.

Instructor: Yoh Kawano
Workshop materials: https://github.com/yohman/workshop-remote-sensing

Satellites are circling our planet, allowing us to "sense" things about the Earth. It is the art and science of making measurements using sensors. Remote sensing has thus become a valuable tool in research and applications in a wide range of disciplines, such as engineering, geology, geography, urban planning, public health, archeology, environmental studies, disaster research, forestry, and agriculture.

In this workshop, you will:

- learn the basic principles of remote sensing methods in research
- have a hands-on session on using Jupyter Notebooks to code in Python
- import Landsat imagery using Google Earth Engine's Python library
- conduct a basic NDVI (Normalized Difference Vegetation Index) analysis

Instructor: Yoh Kawano

In this hands on workshop, you will be provided with a Jupyter Notebook workspace to import census data, explore, clean, and prepare the data for analysis using python. You will also learn how to create compelling choropleth maps with selected variables that can be used for publications.

Course materials:

https://github.com/yohman/workshop-census

Workshop materials available here:

https://github.com/yohman/workshop-python-spatial-stats

Visual interpretations are meaningful ways to determine spatial trends in our data. However, underlying factors—such as inconsistent geographies, scale, data gaps, overlapping data—have the potential to produce incorrect assumptions, as valuable information may be conveniently hidden from the visual output.

One way to address this issue is to amend your visual output with geo-statistical validation. In this workshop, we will use Python to look at one such approach: Spatial Autocorrelation.

Instructor: Yoh Kawano
Course materials: https://github.com/yohman/getting-started-with-gis

In this hands on workshop, you will be provided with a Jupyter Notebook workspace to import census data, explore, clean, and prepare the data for analysis using python. You will also learn how to create compelling choropleth maps with selected variables that can be used for publications.

Workshop material: https://github.com/benjum/idre-data-sci-with-python

Part 1: Introducing data science with python using the iris dataset to cover topics for getting, cleaning, and visualizing data.

libraries to include: openpyxl, csv, json, requests, urllib, pandas, matplotlib, seaborn, plotnine, plotly, altair

Instructor: Yoh Kawano
Course materials: https://github.com/IDREsandbox/gisworkshop

As members of an academic community with a myriad of research initiatives, the need to map information, to spatially analyze, to geoprocess, and to visualize space and time is becoming ubiquitous. But how do you get started? The answer to this question largely depends on what you want to map, and how you want to create your map. The what is usually easy. It is driven by your passion to pursue a research question and to have maps assist in the development of an argument. The how is the difficult part. It may depend on what data you have, what tools you have access to, how comfortable you are with software packages, if you are open to write code or use the command line. Often times, your research question lends itself to a spatial argument. However, providing insights to questions using spatial visualization tools is a process that involves any number of factors, including: data acquisition, data cleanup, geo-enabling data, geocoding data, georeferencing data, visualizing spatial data, overlaying other spatial data, conducting spatial analysis and/or geoprocessing, analyzing results, visualizing results, and publishing results.

In the hands-on segment of the workshop, we will be using the following tools to learn how to analyze and map COVID-19 data:

GitHub
Python
Jupyter Notebooks

Instructor: Yoh Kawano

Workshop materials: https://github.com/benjum/idre-data-sci-with-python

Part 2: A Data Science case study: 3D Mapping of Los Angeles arrest data

- libraries to include: sodapy, pandas, plotly, keplergl

Machine learning with iris dataset

- examples with pytorch and tensorflow

Introduction to SPSS Seminar given on 10/26/2020

Introduction to SEM seminar originally given on February 22, 2021.

This is the second seminar in a three-part series.

1. Confirmatory Factor Analysis (CFA) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rcfa/

The first seminar introduces the confirmatory factor analysis model, and discusses model identification, degrees of freedom and model fit.

2. Introduction to Structural Equation Modeling (SEM) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rsem/

The second seminar explores structural equation models which is an umbrella term that encompass linear regression, multivariate regression, path analysis, CFA and structural regression.

3. Latent Growth Models (LGM) and Measurement Invariance with R in lavaan
https://stats.idre.ucla.edu/r/seminars/lgm/

The third seminar introduces latent growth modeling and how it relates to hierarchical linear models (HLM) and 2) measurement invariance in CFA and how to compare model fit between invariance models.

Confirmatory Factor Analysis in R with lavaan workshop given at UCLA on May 17, 2021 by Johnny Lin, Ph.D.

This is the first seminar in a three-part series.

1. Confirmatory Factor Analysis (CFA) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rcfa/

The first seminar introduces the confirmatory factor analysis model, and discusses model identification, degrees of freedom and model fit.

2. Introduction to Structural Equation Modeling (SEM) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rsem/

The second seminar explores structural equation models which is an umbrella term that encompass linear regression, multivariate regression, path analysis, CFA and structural regression.

3. Latent Growth Models (LGM) and Measurement Invariance with R in lavaan
https://stats.idre.ucla.edu/r/seminars/lgm/

The third seminar introduces latent growth modeling and how it relates to hierarchical linear models (HLM) and 2) measurement invariance in CFA and how to compare model fit between invariance models.

Workshop given on August 16, 2021 by Johnny Lin, PhD
UCLA OARC IDRE Statistical Consulting

This is the third seminar in a three-part series.

1. Confirmatory Factor Analysis (CFA) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rcfa/

The first seminar introduces the confirmatory factor analysis model, and discusses model identification, degrees of freedom and model fit.

2. Introduction to Structural Equation Modeling (SEM) in R with lavaan
https://stats.idre.ucla.edu/r/seminars/rsem/

The second seminar explores structural equation models which is an umbrella term that encompass linear regression, multivariate regression, path analysis, CFA and structural regression.

3. Latent Growth Models (LGM) and Measurement Invariance with R in lavaan
https://stats.idre.ucla.edu/r/seminars/lgm/

The third seminar introduces latent growth modeling and how it relates to hierarchical linear models (HLM) and 2) measurement invariance in CFA and how to compare model fit between invariance models.

Introduction to the analysis of complex survey data in Stata 17.

This workshop describes how to conduct various types of mediation analyses using the PROCESS macro in SPSS.

Workshop discussing ways to present statistical results in peer-reviewed journal articles.

Workshop teaching basic usage of the ggplot2 package in R.

Workshop teaching basics of R Markdown.

This workshop is an introduction to Mplus 8.8. Topics covered include how to get data into Mplus, the basics of Mplus syntax, and how to run basic models in Mplus.

This workshop goes beyond the basics of SPSS syntax so that researchers can accomplish data management tasks via SPSS syntax. Most of the data management commands in SPSS are discussed.

This workshop demonstrates the use of the PROCESS macro in SPSS to graph interaction terms. You will learn how to interpret the coefficients of the interaction terms and the lower-order effects, how to probe the interaction and how to graph it.

Introduction to survival analysis in R using the 'survival' package.

Interactive workshop introducing R.
Webpage: https://stats.oarc.ucla.edu/r/seminars/intro/

Analysis and Visualization of Interactions in R
Webpage: https://stats.oarc.ucla.edu/r/seminars/analysis-and-visualization-of-interactions-in-r/

In this workshop we will discuss missing data in R and introduce some useful packages in R for handling missing data. In particular, we will focus on multiple imputation and how to perform it using the R package, mice: “Multivariate Imputation by Chained Equations”.

In this seminar we discuss generalized linear models and why and when we need to use them. We will discuss several generalized linear modes such as logistic, Poisson, and negative binomial and how we run them in R.

How to create a SAS profile and launch SAS OnDemand.

This workshop introduces the ideas used in meta-analysis, including how to collect and organize data, how to run a basic meta-analysis, and how to interpret the output.

In this workshop we will go over Multinomial and Ordinal logistic regression with examples in R.

Workshop teaching basic Stata usage

This video is private.

This video is private.

Abstract: Urban infrastructure encompasses diverse systems, from residential buildings to critical facilities like power and transportation. Structural Health Monitoring (SHM) provides a framework to ensure the safety and continuity of these infrastructure systems by distinguishing between non-damaged and damaged states. The traditional methods for monitoring the health of such infrastructure face challenges in terms of scalability and generalizability. Emerging approaches, including Generative Adversarial Networks (GANs), Contrastive Learning, and Transfer Learning (TL), combined with SHM domain knowledge, address these limitations by enabling SHM models to function effectively without requiring observed damaged states. The presentation will cover these advancements, including research endeavors in the Taciroglu Research Group (TRG) at UCLA, that pave the way for scalable and reliable SHM in complex urban environments.

Speaker: Hesam Soleimani, Ph.D.
IDRE Postdoctoral Fellow
Civil Engineering
University of California Los Angeles

The panel is led by the following new IDRE fellows (2024-2025) from various disciplines across the UCLA campus.

Barbara Lopez-Doriga, Mechanical and Aerospace Engineering
Claire Schollaert, Environmental Health Sciences
Eulanca Liu, Radiation Oncology
Mo Zhou, Mathematics
Hesam Soleimani, Civil Engineering
Tsukasa Yoshinaga, Head and Neck Surgery

IDRE fellows initiated the event with introductions and lightning talks, followed by Q&A and discussion. They explored how AI technology revolutionizes research methodologies and allows for analyzing and predicting complex datasets, fueling innovation in fields ranging from medical science to engineering and mathematics. The panel also discussed how machine learning harnesses high-performance computing (HPC) resources to process large datasets.

Speaker: Fei-Man Hsu, Ph.D.
IDRE Fellow
Department of Molecular, Cell, and Developmental Biology
University of California Los Angeles

Date: September 27, 2024

Abstract: DNA methylation is associated with pathogenesis and disease progression. The IMPACC cohort includes hospitalized COVID-19 participants with five respiratory illness trajectory groups (TGs) according to the prognosis during acute infection (first 28 days). We report the longitudinal DNA methylation profiling of PBMC from 75 participants of IMPACC cohort in UCLA and find (1) DNA methylation is associated with respiratory status and days post-admission, (2) T cell depletion and band neutrophil backlog were observed during acute infection, and (3) Baseline DNA methylation profiles bear prediction power of follow-up risk course. The characterized CpG loci function in immune cell differentiation and STAT protein phosphorylation suggests an epigenetic-regulated SARS-CoV-2 pathogenesis.

Are you interested in the connections between people? Places? Concepts? Things? Do you want to visualize data in a way that is comprehensible to a human? Do you want to see how information flows through systems? Then network analysis is for you! This introductory workshop will introduce participants to network analysis, with an emphasis on its use for data-driven humanities research. Upon completion of the workshop, participants will be familiar with important concepts and trends in network analysis, be able to construct a basic network from a spreadsheet, run common network analytics, and visualize the results. No prior knowledge of network analysis or programming is required or expected.

This is the slide presentation for UCLA OARC/IDRE Workshop "Learning Scikit-Learn: Basics" in Summer 2021.

The pdf slide and the demo files can be downloed from the github repo: https://github.com/huqy/learning-sklearn

Video Chapters:
00:00 Opening remarks
03:25 Knowing scikit-learn in minutes
12:21 Simplified workflow for a machine learning project
15:09 Data input, loader and generator
20:20 Splitting the data
24:51 Scikit-learn core objects
29:11 Data pre-processing
32:08 Models and meta-estimators
43:35 Model persistence
48:29 Scikit-learn extension libraries
53:32 Key Takeaways

Instructor: Yoh Kawano

Workshop materials: https://github.com/benjum/idre-data-sci-with-python

Part 2: A Data Science case study: 3D Mapping of Los Angeles arrest data

- libraries to include: sodapy, pandas, plotly, keplergl

Machine learning with iris dataset

- examples with pytorch and tensorflow

Workshop material: https://github.com/benjum/idre-data-sci-with-python

Part 1: Introducing data science with python using the iris dataset to cover topics for getting, cleaning, and visualizing data.

libraries to include: openpyxl, csv, json, requests, urllib, pandas, matplotlib, seaborn, plotnine, plotly, altair

Monday, Feb 10, 2025, 10:00 am - 12:00 pm

Graph databases are at the frontier of data science, humanities, and AI computing. Built to handle dense, highly interconnected data, graph databases underpin many social networks, big data applications, and new research initiatives, but are still not as widely understood or deployed as traditional relational databases.

Building off the Introduction to Databases (Part 1) workshop, this session introduces graph databases, with a focus on Neo4j and Wikibase. Spanning two hours, the first half of the workshop will focus on graph database fundamentals, the network data model, and the differences between graph and traditional SQL relational databases.

We will then demonstrate common tasks and procedures in Neo4j and Wikibase, and show how to use graph databases as a backend to data visualization and manipulation in python / juptyter notebooks.

This workshop assumes that participants have general knowledge of relational databases or previous participation in part 1 of the database series. Participants will need to install a working copy of Neo4j Desktop Community Edition by visiting https://neo4j.com/deployment-center/, scrolling to the section Neo4j Desktop, then download and install the program.

Pre-workshop Instructions:

Participants will need to install a working copy of Neo4j Desktop Community Edition by visiting https://neo4j.com/deployment-center/, scrolling to the section Neo4j Desktop, then download and install the program.

Instructor: Ryan Horne

Databases are central tools for organizing research. In addition to storing data, databases allow for data to be quantified in meaningful ways. The process of creating a database can also help clarify relationships within content. Relational databases are designed to reduce redundancy, increase efficiency, and decrease errors.

This introductory workshop on databases offers participants a foundational understanding of database design, relational structures, and key technical concepts. Over two hours, attendees will learn the essentials of organizing and managing complex data, exploring primary and foreign keys, table relationships, and normalization to reduce redundancy and errors. The workshop will introduce popular database platforms, such as MySQL, PostgreSQL, and MongoDB, as well as beginner-friendly tools like Google Sheets, providing insight into selecting the right tool for different projects. Through hands-on activities and discussions on ethical data use, participants will leave equipped with practical skills and resources to begin building and querying databases effectively. No prior experience is required.

October 9, 2024 @ 11:30 am - 1:30 pm

Are you interested in creating and publishing 3D virtual environments on the Web? Would you like to be able to quickly spatially model multimedia assets? This workshop will introduce you to foundational techniques and resources which will enable you to produce WebXR (eXtended Reality) content.

This spring, Mozilla sunsetted their WebXR platform, Mozilla Hubs, and builder, Spoke. While the company did release the platform as an open-source tool available at the https://hubsfoundation.org/, personal installations locally or on the cloud can present a challenge with the current level of documentation and community support.

This workshop will cover two new and exciting online alternatives to Hubs. Mud Verse (MUD VERSE – MUD Foundation INC / Art + Tech + Community) is supported by “an innovative nonprofit organization dedicated to the exploration of art, technology, and education” and allows for Spoke files to be imported. The platform operates nearly identically to Mozilla Hubs and it is free to register an account. A second option is 8XR – an HTML5 native game engine. As a game engine, published environments can have increased interactivity built-in and added through additional programming.

During the workshop, participants will be shown how to:

- How to find 3D resources to build a project quickly
- Import, augment, and arrange 2D and 3D visual assets
- Integrate text, video, and audio materials into the WebXR platforms
- Publish and share virtual environments
- Learn how to navigate within the published XR spaces and participate interactively with other digital avatars

No prior technical experience necessary. All levels welcome.

Slides:

By Benjamin Winjum, Project Scientist, OARC
December 7, 2023, 10-11AM via Zoom

The application of machine learning (ML) and high performance computing (HPC) to humanities research holds tremendous potential, but there are significant challenges to developing a community of experts that can work together at this intersection of topic and technique. HPC experts may not have the background necessary to appreciate how their tools can best be brought to bear on the diverse datasets and conceptual approaches of humanities researchers, and humanities researchers may not have the well-honed background in computational and statistical modeling that resides with ML or HPC specialists. We will approach this topic by focusing on a few case studies in the humanities that illustrate how ML and HPC can both be used to powerful effect, whether by streamlining the initial processing of massive data, modeling and visualizing humanities-relevant datasets, or using deep learning frameworks and the speed and power of GPUs to rapidly prototype predictive models.

Information systems since the advent of the World Wide Web have offered increasingly sophisticated and reliable tools for inquiry and scholarship based upon centuries-old practices of annotation. What are the connections and tensions between visual and verbal media? How do idiosyncratic readerly practices of scribbling in margins of a book translate into W3C standards for networked multimedia scholarly communication, and what impacts will community standards development have upon the future of knowledge production (or your career)?
This workshop will provide a theoretical introduction to the ancient and still evolving social practice of annotation and situate leading ideas in the concrete tools and practices of digital knowledge production. Taking a 3-tiered approach (simple, intermediate, complex), it will present several handy tools for immediate and long-term application.

This workshop is part of the Research Collections and Digital Scholarship series — a collaboration between the Office of Advanced Research Computing (OARC) and the UCLA Digital Library Program.

Are you interested in the connections between people? Places? Concepts? Things? Do you want to visualize data in a way that is comprehensible to a human? Do you want to see how information flows through systems? Then network analysis is for you! This introductory workshop will introduce participants to network analysis, with an emphasis on its use for data-driven humanities research. Upon completion of the workshop, participants will be familiar with important concepts and trends in network analysis, be able to construct a basic network from a spreadsheet, run common network analytics, and visualize the results. No prior knowledge of network analysis or programming is required or expected.

In this presentation, we will explore how the HPC can be used to accomplish common DH tasks, such as network and GIS analysis with large data sets which would be computationally difficult or expensive on typical desktop / laptop hardware. We will use least-cost pathing with GDAL, network analysis with Python, and general DH data analysis and visualization with Jupyter Notebooks. This workshop is designed to highlight the usability and accessibility of these common tools in the HPC environment.

Most would agree that in terms of creating virtual experiences as part of research, more context is better. Contextual information allows for a bigger picture to be seen and can provide in situ understanding, despite being remote. With the proliferation of 360 cameras, creating immersive virtual tours has become more commonplace. Whether it is to document the layout of a hydroelectric power plant, provide archival information about historical relics, such as the HMAS OVENS submarine, or share information regarding a cultural heritage site, such as Google Arts and Culture’s Street View tours, immersive 360 capture with annotations has become a strong visual research method and product.
In addition, many fields of research look to present information or scenarios and then survey a sample of participants regarding their experience. Interactive surveys or modules can provide valuable research data for any number of disciplines.

In this workshop, participants will be introduced to Adobe Captivate, a software for building e-learning interactives and annotated 360 virtual tours. Focusing on examples that bridge UCAL’s Biomedical Library space and digital collections, participants will learn how to build a virtual tour with 360 images and video. They will also learn how to annotate their tours with collections materials and how to build some of the platforms various interactive activities, which include:

Multiple choice
True/False
Fill-in-the-blank
Short Answer
Matching
Hot Spot
Sequence
Rating scale
Random question

No prior experience with Adobe products will be needed. However, this product is not part of the Adobe Creative Cloud Suite. If participants want to follow along during the workshop, they will need to purchase and download Adobe Captivate before the workshop.

This workshop is part of the Research Collections and Digital Scholarship series — a collaboration between the Office of Advanced Research Computing (OARC) and the UCLA Digital Library Program.

Within the scholarship of teaching and learning, it has been proven time and again that “Assembling and curating specimen collections is a valuable educational exercise that integrates subject-specific skills such as field collection, curation, identification, organization, and interpretation of relationships.” (Lucky et. al 2019) Digital collections increasingly avail students and scholars of primary source research materials that were only accessible previously by exclusive arrangement in person, if at all. Rare and unique items, such as manuscript documents, cultural artifacts, and historical print ephemera, still present methodological challenges for intensive engagement, even when they are readily and openly accessed. In this third workshop of the series, the focus will shift to strategies for the design, development, and implementation of curriculum-based learning activities that engage collections of IIIF (International Image Interoperability Framework) content.

In this workshop, participants will:
- Learn various strategies for learning activity design using curated collections of items
- Create usable sample learning activities in Bruin Learn, using IIIF-based collections content and embedded IIIF-compatible tools

All levels are welcome. This workshop series is a collaboration between the UCLA Digital Library Program and the Office of Advanced Research Computing. Additional leadership provided by UCLA faculty who have been participating in the LMS Design Lab.

References
Lucky, A., Branham, M. & Atchison, R. Collection-Based Education by Distance and Face to Face: Learning Outcomes and Academic Dishonesty. J Sci Educ Technol 28, 414–428 (2019). https://doi.org/10.1007/s10956-019-9770-8

Title: Teaching & Research w/ Digital Collections (2): Tools and Practices for Engaging Digital Collections Content

Date: 1/26/22, Time: 1-2:30pm PT

Description:

The active engagement of digital primary source collections content can, and often does, go beyond viewing materials on collections websites. For sustained inquiry, a wide variety of tools and platforms have been developed that support in-depth scholarly analysis of high quality, reliably-hosted image-based digital content and enable the creation of public-facing scholarly work products. This workshop is the second in a series about teaching and research with digital collections, which highlights the affordances of IIIF content (for International Image Interoperability Framework) but it does not require prior knowledge or skills from the first workshop. It will focus upon the “interoperable” use of digital materials outside of their original institutional context, in both lightweight, easy-to-use tools for analysis, presentation and annotation (such as Storiiies, Exhibit.so, Mirador, and MISE), and more robust platforms for digital scholarship, such as Scalar.

In this workshop, participants will learn:
- What are IIIF manifests and image URLs, and how can they be used in IIIF-compatible tools and platforms?
- How to extract details from image-based content
- How to import materials as linked content into various platforms
- How to annotate, caption, and present content in multimedia essays and exhibits

All levels are welcome. This workshop series is a collaboration between the UCLA Digital Library Program and the Office of Advanced Research Computing. Additional leadership provided by UCLA faculty who have been participating in the LMS Design Lab.

Teaching and Research w/ Digital Collections (Part 1): Navigating A World of High-Quality Primary Source Materials
Thursday, Jan 19, 2023 - 1:00pm to 2:30pm
Online and in-person

Scholarship of visual materials often requires close looking and comparison across works. Digitization programs at “GLAM” institutions (Galleries, Libraries, Archives, and Museums) around the world, including the UCLA Library, have resulted in extraordinary collections of digital facsimiles of rare and unique primary source books, documents, artifacts, and ephemera, available online. For scholarship and curricular research assignments, sophisticated digital viewers present high-resolution image-based content, which allows for close and comparative examination of materials. Interoperable content, which follows certain standards of web-delivery, has become crucial for institutions for contrasting objects from different collections. Increasing stability in digital visual scholarship means working from authority files from museums, libraries, and archives — institutions are committed to long term preservation and standards. Finding items of interest, however, can be a challenge, as digital collections are not universally searchable using familiar tools and practices, and they often differ between institutions.

This workshop will introduce participants to International Image Interoperability Framework or IIIF (generally pronounced “triple-eye-eff”), which is a way to standardize the delivery of images and audio/visual files from servers to different environments on the Web. In this first workshop session in a series of three, participants will learn:

- How to locate and use digital collections with IIIF content
- The benefits of IIIF materials for research and teaching
- Image API
-Presentation API

All levels are welcome. This workshop series is a collaboration between the UCLA Digital Library Program and the Office of Advanced Research Computing. Additional leadership provided by UCLA faculty who have been participating in the LMS Design Lab.

Workshop will be held both:

in person at the IDRE Portal (520 Portola Plaza, 5628 Math Sciences Building, Los Angeles, CA 90095)
over Zoom

Recorded: Tuesday, October 25, 2022 - 12:00pm to 2:00pm

Description
Whether it is interviews, documentary coverage, or contextual footage, video has become a common practice within research. This course is designed to help scholars manage and make use of their research video footage. The workshop will discuss various best practices regarding file management, editing platforms, and techniques for putting footage together to meet your research goals. Participants will be introduced to Davinci Resolve to learn the basics of a preliminary editing process with industry standard software. Software alternatives, such as Adobe Premiere and iMovie will be discussed. In addition, related resources will be shared so that participants can continue to progress in the video editing process.

Any questions about this workshop can be emailed to Francesca Albrezzi at falbrezzi@ucla.edu.

Slides: https://docs.google.com/presentation/d/1JTDgoJoh5vMdxkd6ySOIk26DdpDGc6twpOtUPSl3r8Q/edit?usp=sharing

NFTs seem to be everywhere now. NFT stands for non fungible tokens. The creation and exchange of NFTs has been revolutionary in many ways, particularly in terms of arts and services markets. For example, in the art market, NFTs use blockchain technology to allow for the secure transfer and proof that an original piece of digital art has been sold and is now in the possession of the buyer. It also allows for what we in art history may refer to as provenance of an artwork to be more carefully tracked. The benefits of blockchain is that it enables a global secure transfer of private smart contracts between parties that are stored on an open decentralized network. The result has exposed art market sales more openly and shifted the value of digital art drastically.

In this workshop you will learn about rarity, royalties, and crypto wallets. We will look at some examples of the way NFTs are being used in different markets, but will focus most specifically on growing digital art markets. You will also learn how to set up your own crypto wallet, create your own NFT, and begin mining Ethereum, a popular crypto currency, which is used in the buying and selling of many NFT artworks and items.

No prior knowledge or experience required.

Any questions about this workshop can be emailed to falbrezzi@oarc.ucla.edu.