The final exam will come out at midnight on Tuesday, August 2.
It will remain available until 11:59pm on Wednesday, August 3.
You can "Fetch" the exam anytime during that interval. Once you click "Fetch", you have 3 hours to submit to JupyterHub (unless otherwise specified).
Anything we've covered this semester is fair game, from the first day to the last.
That said, the mechanics will focus more on the second half: probability, statistics, linear algebra, and machine learning. Doing these things arguably requires skills acquired in the first half.
Same as the midterm: a Jupyter notebook, hosted on JupyterHub.
Some portions of the final will be autograded, but most won't be.
Unlike the midterm, the questions in the final exam are very much cumulative: that is, the later parts of the questions build on the previous parts. This is intentional: data science pipelines very much rely on long-term, iterative development that builds on early contributions.
Same as the midterm: please don't use the internet. Things that are ok to use:
Obviously I can't directly observe while you're taking the final, so it's honor system.
With respect to Slack, I will be available to answer technical questions you run into on the final. I will try to answer questions as they come, but I will explicitly make myself available on Tuesday from 9am to 5pm and on Wednesday from 11am to 2pm. Outside those ranges I'll respond as quickly as I can, but no guarantees.
Sweet, sweet vengeance! Also a phenomenal opportunity to help improve the course.
This is the first time CSCI 1360 has ever been taught, so I appreciate everyone's patience this semester as we worked out the wrinkles. If you have any suggestions for future iterations of the course, please use the link below!
If you have any questions, don't hesitate to ask on Slack! That's what it's for.