2021 Summer boot camp

Videos and Powerpoints will be posted after the event.

RShiny and PyFlask: Building Interactive Web Apps for Data. Building data dashboards accessible via the Internet is an excellent way to allow users to interface with data using tools they are already familiar with. This tutorial will teach participants how to build their own web-based data dashboards using RShiny (in R) and Flask (in Python). We will begin with building a simple example of each tool, then move to advanced applications of these packages.

NumPy, Pandas, and Scikit-Learn: Prediction with Decision Trees. Numpy and pandas are the foundation of the python data science stack: most python machine learning libraries utilize their objects and data structures, including scikit-learn. Come learn the basics of numpy and pandas, and learn how to build and train decision trees for classification.

An Introduction to Machine Learning for Materials Science: A Basic Workflow for Predicting Materials Properties. This tutorial will introduce core concepts of machine learning through the lens of a basic workflow to predict material bandgaps from material compositions. As we progress through this workflow we will highlight key steps, challenges that can come up with materials data, and potential solutions to these challenges. The core workflow we will introduce includes Data Cleaning, Feature Generation, Feature Engineering, Establishing Model Assessment, Training a Default Model, Hyperparameter Optimization, and Making Predictions. By the end of the tutorial I hope that you will have a better understanding of these core concepts, and how they can all fit together. If you want to preview the materials ahead of time you can find them on Nanohub here: https://nanohub.org/tools/intromllab

The Materials Simulation Toolkit for Machine Learning (MAST-ML): Automating Development and Evaluation of Machine Learning Models for Materials Property Prediction. This tutorial contains an introduction to the use of the Materials Simulation Toolkit for Machine Learning (MAST-ML), a python package designed to broaden and accelerate the use of machine learning and data science methods for materials property prediction. Through hands-on activities, we will use MAST-ML to (1) import materials datasets from online databases and clean and examine our input data, (2) conduct feature engineering analysis, including generation, preprocessing, and selection of features, (3) construct, evaluate and compare the performance of different model types and data splitting techniques, and (4) conduct a preliminary assessment of model error analysis and uncertainty quantification (UQ). MAST-ML Tool Github page: https://github.com/uw-cmg/MAST-ML

The Informatics Skunkworks: A Program for Undergraduate Research at the Interface of Data Science and Materials Science and Engineering.In this presentation, I will go over the new infrastructure and ecosystem that we are developing for the engagement and training of undergraduate students (UGs) in research at the interface of data science and materials science and engineering, with a focus on the use of applied machine learning (ML) in materials informatics. I will describe the resources that we have developed to lower barriers to starting research projects, including (a) curriculum to train UGs in relevant data science and materials informatics, (b) software tools that augment existing ML packages to be UG accessible, and (c) authentic and appropriate-level research problems.

Crossing Darwin and Computer Science: The Staying Power of Evolutionary Algorithms. Beyond revolutionizing our views on life and the world in which we live, Darwin's theory of evolution has been the basis and ongoing inspiration for an entire branch of machine learning. Evolutionary algorithms are frequently used to tackle some of Computer Science's most nefarious challenges'the notorious NP-complete problems'in applications as varied as mirrors designed to funnel sunlight to a solar collector, antennae designed to pick up radio signals in space, walking methods for computer figures, and optimal design of aerodynamic bodies in complex flowfields. In this tutorial, Dr. Bodily will lay out the theory behind genetic algorithms, illustrate several applied examples of genetic algorithms from his research and other real-world applications, and will involve participants in an interactive, live-coding demo to implement a genetic algorithm that can be repurposed for a variety of applications. Come prepared for a fun, engaging, rewarding learning experience!

Data Science Research Discussion Panel: Tools Applications, Networking, and Collaboration: Join us to hear our presenters share their experience with the data science tools they use in their research, their plans for future projects and grants, and how they recommend students continue growing their skills in these areas.