On Friday, June 23rd, SAS and Women in Big Data hosted a workshop for 20 attendees, ranging from students to professionals. The workshop focused on data analysis and machine learning using SAS’s Visual Analytics tool and Python API. The workshop was kicked off with an insightful presentation by Jenn Mann, SAS Executive Vice President who flew in all the way from North Carolina.
Jenn Mann is an industry leader well-known for her skills in creating a strong positive workplace culture filled with talent and diversity with room of growth. She addressed the unbalanced proportion of women and men in tech companies, and even worse proportions in senior leadership and technology roles in Silicon Valley. The primarily reported issues are that women in tech companies experience an unwelcome work environment, a sense of isolation, difficult work life integration, and lack of advancement.
At SAS, Jenn takes pride in the fact that the company, as pioneers in the workplace culture, has a stronger proportion of women throughout the company than most technological companies along with a strong retention. She encourages organizations and individuals to encourage companies to improve workplace cultures to help bring and retain women in the workforce. She shared some key components that SAS follows to create a strong workplace culture, two of which are:
After Jenn’s presentation, Stacey Syphus, a senior manager and instructor at SAS, taught a workshop on data analytics using SAS’s Visual Analytics tool. Supplied with a very handy set of course notes, attendees were provided several ways to import, manipulate, and display a large course-supplied data set. Working with graphs ranging from maps to bar charts and time series plots, we were then able to use SAS Visual Analytics to create and customize an interactive report. It was a well thought-out workshop which successfully showed how versatile and powerful the tool could be.
Following Stacey’s workshop, Nadeem Chaudhry, a SAS senior solutions architect, taught a workshop on machine learning. With an introduction on the various machine learning algorithms (supervised, unsupervised, and semi-supervised), attendees followed a Jupyter notebook to perform a supervised machine learning algorithm. Armed with the SAS Python API, we learned how to:
After determining that gradient boosting was the desired model, in a great teaching moment, Nadeem introduced an additional (later discovered to be better) model, a random forest model. The takeaway: it is important to have a baseline model to then continue the search for even better models.
Github link to code for the workshop: https://github.com/datawicked/PythonExamples
After the half day of training was complete, SAS continued to reward attendees with a raffle where two attendees won e-learning seats in the SAS Academy for Data Science! Congratulations to everyone on a successful workshop!