CSE-6040 - Computing For Data AnalyticsAssignment Due: Thursday, Dec 5th @ 1159 UTC
Peer-grading Due: Wednesday, Dec 11th @ 1159 UTC
Two deadlines = Two steps: You must submit a project AND complete peer grading in order to receive credit. If you miss either one of those things you will not get credit.
Point Value: Up to 3% added on to your final grade
Grading: Peer-graded (brief rubric below)
Where/how to submit work: In the Extra credit assignment. You can find it linked towards the bottom of Canvas/EdX titled "Extra Credit". Complete your work there and hit "submit".
No Plagiarism: This will result in a 0 on the extra credit, and a report to GT OSI
Assignment: Pick a topic that interests you. Find some data related to that topic and come up with some questions you might answer using that data. Create a Jupyter notebook that processes the data and implements an analysis to address your questions. We expect your notebook will need to clean the data, analyze it using tools you’ve learned in this class, and produce some outputs that will help you show what you’ve learned from the data. Remember that your outputs may disprove your original theory or be inconclusive, a situation that is very common in analytics; this result is okay if you demonstrate a solid effort to answer your questions.
Below are the specific requirements for your project. Outside of what's required below, the project is open-ended and other design considerations are up to you.
These requirements are strict and you may receive a 0 if your project does not follow them.
import <package>pd.read_csv('test_data_1.csv') is all you need. Do not use os.getcwd() or anything to generate a dynamic path. It won't work for grading. Simply read in the file directly like above The section below applies to you as a person submitting a prjoect, then to you as a peer grader.
Can you provide some example projects? The past and current midterms closely resemble the style we’re looking for, but that level of complexity is not necessary (but is doable!). In particular, look at how each one starts out with some data, goes through some cleaning and analysis, and then outputs something, whether it’s a specific output or just well-documented insights.
How long should I spend on this? We can’t answer that for you, but it’s reasonable to expect this to take 10-15 hours, depending on how comfortable you are with the material we’ve covered in the class so far. We’re not asking you to do anything new, just use things you’ve learned in the class.
How do I upload the dataset or files to Vocareum? Open up the assignment from Canvas/edX like a normal notebook. When you're in the folder view (click the Jupyter logo at the top left if you're in a notebook), you can click the "upload" button at the top right to upload files.
Can the deadline be extended for this? Unfortunately, no. We need to leave enough time for the teaching staff to calculate and input grades before the Georgia Tech grade submissions deadline.
I had a great idea and a good plan, but once I did the analysis the results weren’t clear. That’s ok! As long as you explain all of that in your notebook, you will still be eligible for credit. A well-documented project and breakdown of your findings (or lack thereof) is still useful. Academic research papers and real business projects often end with this result, it’s just part of the process.
The dataset I want to use can’t be publicly shared. Please choose a different project or dataset to use. Your peers need to be able to run your code against the data to see the results. You can investigate using a smaller piece of the dataset, or redacting some of it, but we recommend you don’t spend much time on that step. You’d be better served by finding a new dataset and focusing your time on analyzing it and building your project around that.
Can I create the notebook locally and then upload it? Yes, but your code must be executable within the Vocareum/Jupyter environment. The file will need to correctly link to the data file once uploaded, and you’ll need to make sure that any package differences don’t affect your work. We won’t be able to help you debug this, so you will need to make sure this causes 0 issues before the deadline.
Will there be partial credit? Grading will be done by your peers and may result in any number score in the range [0,3]. Everyone should be able to get either 2-3 points with enough effort.
I made a mistake in the in my submission, can I resubmit it? If it’s before the assignment deadline, then yes you can submit as many times as you want. When the deadline hits, your peer reviewers will be given only your latest submission.
Can I collaborate with another student? Yes, but each student will need to submit their own work. You can share a dataset and work together to brainstorm methodologies, but your analysis must be your own. Please also name the student you worked with in the top block of your notebook.
How do I access and submit my peer reviews? The peer review process is accessed through the same link in your LMS (Canvas or EdX). The day after the assignment closes (usually by noon ET the following day), you simply click on that link again and it will take you to an interface to view your peer submissions and submit grades + feedback.
My notebook returns a warning, is that ok? Errors are instant 0’s, but warnings are allowed. However, you should ensure that the warning you’re getting is not related to something you’re doing incorrectly. Sometimes they’re notifications of deprecated functionality, sometimes they help highlight a way you incorrectly referenced data.
Can my code import the data from a website instead of storing it in Vocareum? No, the data must be uploaded into the Vocareum environment for peer grading.
One of my peers gave me a 0/3 and I think I deserve higher, can you regrade it? No, there will be no re-grades.
The peer feedback window has an option for “recommend for future exam use”, what is this for? That’s a flag we ask you to enter [0,1] for where 1 indicates you think that assignment would be good for the instructors to review and use as inspiration for a future exam.
Can the TA’s take a look at my project to confirm it was submitted correctly? As long as you followed the instructions above, you should be good to go. If you want the TA's to look then please make a private piazza post, but depending on timing we may not be able to reply quickly (if you ask two days before the deadline, we may not get a chance to look until after the deadline hits)
Here’s how to get an answer: * If it’s a general question: ask as a follow-up in the Extra Credit Piazza post * If you want to brainstorm with others and get suggestions: create a public piazza post or ask in Slack. You can work with others to refine your ideas or get advice, just make sure the work in your notebook is your own. * If it’s specific to your project: create a private post but please use your judgement—we can’t guarantee that we’ll respond in a timely fashion, and while we would like to help everyone out with tips on their projects, we simply don’t have the ability due to end of semester activities. We would prefer that you continue making progress on the project while waiting for a reply fr