Overview

Exam 1 will be released on Sunday, May 14 at 4:30pm and must be completed by         . There will be a thiry-minute grace period for students who wait until the last minute to submit on Gradescope. Any submissions received after 5pm ET will receive an automatic 20% deduction.           Submissions will not be accepted after 11:59pm ET on May 16.

The exam will focus primarily on machine learning techniques from lecture 21 (introduction to machine  learning) through . Having said that, it is perfectly reasonable to expect you utilize any techniques and approaches we have learned throughout this class  (e.g. importing data files, cleaning/tidying data, exploratory data analysis, data transformation, data       visualization).

It will consist of data analysis in R/RStudio and be submitted via Gradescope (similar to exam 01.)

Rules & Notes

Students with SDS accommodations

Students who have registered SDS accommodations with time extensions for exams will receive them on exam 3. The extension will apply submission deadline (e.g. someone with 1.5x time extension for exams is required to submit the exam by 4:30pm on Wednesday, May 17).

Cornell University’s Code of Academic Integrity

1. A student shall in no way misrepresent his or her work.

2. A student shall in no way fraudulently or unfairly advance his or her academic position.

3. A student shall refuse to be a party to another student’s failure to maintain academic integrity.

4. A student shall not in any other manner violate the principle of academic integrity.

Other AI information

This is an individual assignment. Everything in your repository is for your eyes only.

You may not collaborate or communicate anything about this exam to  except the instructor. For example, you may not communicate with other students, other TAs, or post/solicit help on the internet, email or via any other method of communication.

The exam is open-book, open-note, so you may use any materials from class as you take the exam. You may make use of online resources (e.g. package documentation, StackOverflow, Google search results)

but you may not directly copy and paste from these sources, but instead you need to adapt the code to fit your specific task. You must explicitly cite where you obtained the code using a code comment #  immediately near the appearance of the reused code in the file. Any recycled code that is discovered

and is not explicitly cited will be treated as plagiarism.

Clarification questions may be sent to the course email account ([email protected]) only. You may

 email the TAs questions about the exam.

Submission

You must submit a PDF to Gradescope that corresponds to the  .qmd file on your GitHub repository in order to receive credit for this portion of the exam.

You must upload a PDF, not HTML. Any non-PDF submissions will  be graded.

Your PDF must be the  generated by Quarto.

Mark the pages associated with each question. If any answer for a question spans multiple pages, mark all associated pages.

Failure to mark the pages in Gradescope will result in lost points. Only pages that are marked will be graded and eligible to receive credit.

Make sure that your uploaded PDF document matches your  .qmd and the PDF in your GitHub repository exactly. Your PDF should be fully reproducible from the  .qmd file.

Grading

  Total: 50 points