Registration for 2018 WiDS C'ville is full. Add your name to the mailing list to receive updates on the event livestream, and future events and opportunities to participate.
As part of the 2018 Charlottesville Women in Data Science (WiDS) Conference, we are offering two skill sessions on key topics in data science.
They will run concurrently from 1:00-2:15 p.m., so WiDS registrants may only attend one. You can select a skills session when you register for the WiDS conference.
Introductory Skills Session: Workshop in R
Instructor: Marieke K Jones, PhD Research Data Specialist, Claude Moore Health Sciences Library, University of Virginia
Description: This workshop is directed towards conference participants who want an overview of the statistical computing programming language R. Complete beginners are welcome!
This hands-on workshop will introduce R in the context of its data manipulation and visualization capabilities. Via interactive demonstrations and hands-on exercises, attendees will learn how to import data, create data summaries, and generate data visualizations. Attendees are asked to bring a laptop with the following 3 things installed: R, RStudio, and the tidyverse package.
Intermediate Skills Session: Workshop in Machine Learning
Instructor: Charlotte Blais, Data Analyst, IBM
Description: Please join members of IBM's Advanced Analytics group for a discussion of linear and nonlinear approaches to predictive modeling. This workshop will touch on aspects of the modeling process including feature engineering, data visualization, and model selection. Class participants will ultimately work in groups to predict the likelihood of individual passenger survival in the notorious 1912 Titanic shipwreck.
Attendees are asked to bring a laptop with RStudio installed.
Intermediate Skills Session: The “Tour de Package” in Machine Learning - How to Build Models in Scikit-Learn, h2o, and Keras
Instructor: Catherine Ordun, MBA, MPH, Booz Allen Strategic Innovation Group, Data Science & Machine Intelligence
Description: In this hands-on tutorial, we will build different supervised classification models using popular Python machine learning packages. We will build a model using the Scikit-Learn package showing participants how to partition data, fit, predict, and evaluate a model. From there we will build another classification model in the h2o package that offers many evaluation metrics to interpret model performance. Finally, we will build a deep learning model in Keras, a user-friendly API that abstracts languages like TensorFlow, making the syntax easy to interpret similarly to Scikit-Learn. Users are encouraged to bring their laptops with Python 2.7 or 3.x installed. It is highly recommended to have pre-installed through ‘pip install’, Scikit-Learn, h2o, and Keras. The Jupyter Notebook used in the tutorial will be made available after the session.