The homeworks will each have two components:

  • A conceptual component in which you will be asked to read a case study, article, or research paper – about how a data science technique works or doesn’t in practice – and then answer several short answer questions. This part may further include a few short, theoretical (mathematical) questions.
  • A programming component (in Python) in which you will be asked to implement some of the solutions taught in class on either synthetic or real-world data. You may be asked to submit predictions on a test set (e.g., recommendations, or prices at which to sell an item) that will then be scored by the teaching staff.

Attestation at top of every homework

Note the following guidelines and attestations that you will be asked to complete for every homework.

I understand that I have to submit this homework assignment in two places:

  • gradescope – where I will submit a pdf of the assignment and tag each page of my assignment with the problem to which it corresponds.
  • github (via classroom50) – where I will commit and push up this notebook and the rest of my code, in a runnable format.

I understand that I will reach out to the teaching staff via office hours or EdStem at least 5 days before the homework is due if I have any issues submitting the assignment on either of these platforms, such as I am not on the class roster. I understand that I may receive points off (up to full value of the homework assignment) if I do not submit to both locations.

I understand that the teaching staff prefers that I make multiple commits on github (via classroom50) as I am completing the assignment (for example, after I finish each problem), and that this information may be used to assess overall understanding, progress, and qualitative assessment/grading.

I understand that I must fully disclose my usage of AI in the box below and at the top of each problem. I will follow the AI usage policy, and am responsible for everything that I submit. I understand that quizzes and other assessments may evaluate understanding of what I submitted, including asking me to interpret what code does

Homework grading

Each question will be given a score from 0 - 2 points.

  • 0 point: The question was not attempted.
  • 1 point: If you attempt it, but there are either substantial methodological errors or major conceptual misunderstandings of the material
  • 2 points: Essentially correct (may have minor errors), and the conceptual discussion is sufficient.

We expect that most students will receive 2 points on most homework questions. There may also be bonus questions – when calculating grades, the “denominator” for 100% will be (the number of mandatory questions)*2.

Dropping grades and late days

  • We will replace your lowest homework grade with your quiz average (if it is higher). However, we expect that quiz grades will be lower on average than homework grades – we do not recommend that you skip any of the homeworks unless necessary.
  • In addition, you will be allowed four total late days during the semester. These late days allow you to turn in homework up to 24 hours late, with no penalty; you may also use all four late days on a single homework assignment (allowing you to turn it in up to 96 hours late), but that will leave you with no late days for other homeworks. When submitting an assignment late, please mark at the top how many late days are used for this assignment and how many you have used before, if any.

Submitting assignments as pdfs

Due to limitations of gradescope, you have to submit your assignment as a pdf. Some people have reported difficulties with exporting jupyter notebooks as pdfs – it is your responsibility to be able to do this export. Several things that work for us:

  1. Printing to pdf from the browser directly.
  2. Exporting to HTML first, and then opening the generated file in a browser and printing to pdf.
  3. Going through LaTeX to generate a pdf. Visual Studio can directly export notebooks to pdf if you have Pandoc and a LaTeX compiler installed.

Important note: please make sure that your pdf has reasonably sized pages and quality (easily readable from Gradescope interface on a regular laptop screen), and that correct pages are assigned for each question (Gradescope allows you to preview submissions). Please note that for every question that does not satisfy any of these requirements, 0.5 points will be subtracted.