- Quarter 2 (DSC 180B, Winter 2023)
- Quarter 1 (DSC 180A, Fall 2022)
Welcome to the capstone program! The capstone program is a two-quarter sequence (Fall 2022 and Winter 2023) in which you will be mentored by a faculty or industry expert in their domain of expertise. By the end of Quarter 2, you will design and execute a project from that domain in teams of 2-4. You can see the projects from last year here.
At a high level, here’s how the capstone program is organized:
- In Quarter 1 (DSC 180A), you gain background information in your mentor’s domain, by means of replicating a known result. By the end of Quarter 1, you will have completed a replication project (known as the “Quarter 1 Project”) and will have a proposal for a more independent project (known as the “Quarter 2 Project”, or the capstone project).
- In Quarter 2 (DSC 180B), you execute the Quarter 2 Project you proposed at the end of Quarter 1.
Throughout both quarters, there is a “methodology” component that will supplement your knowledge of your domain with best practices in software engineering, project management, and effective communication.
In Quarter 2, you will execute your Quarter 2 Project, the proposal for which you submitted at the end of Quarter 1. You will also prepare to present your work to a variety of audiences, in a variety of formats (most notably, as a poster at our in-person capstone showcase on March 15th).
A few things to note before the quarter begins:
- Sections begin on Monday, January 9th. Confirm the date and time of your discussion section in the Winter 2023 tab of the Domains and Mentors spreadsheet.
- If your section is held in SDSC, follow the instructions found here to enter.
- Make sure to enroll in your section ASAP if you haven’t already, and to start participating from Day 1.
- There will be methodology lessons released each Monday to guide you through your Quarter 2 Projects, along with methodology Office Hours for you to get help. However, there are no live lecture help sessions or methodology assignments.
- Make sure you can access the following sites; email Suraj if you can’t:
As in Quarter 1, methodology lessons in Quarter 2 will be presented as a series of lecture notes that you read outside of class. Unlike in Quarter 1, there are no synchronous “lesson help” sessions, though you are still free to come to office hours to discuss the methodology lessons with the instructor and course staff. Lessons will be released on the Mondays listed in the calendar on the course homepage.
Quarter 2’s methodology lessons are designed to support you as you complete your Quarter 2 Project deliverables (of which there are many). With that said, the time you spend on methodology lessons will be minimal, since most of your time will be spent executing your project.
The live lecture slot on Monday will be used for Suraj’s office hours; the live lecture slots on Wednesday and Friday will not be used for anything.
As in Quarter 1, discussion section will focus on your domain. Each week, a representative from each group will give a “weekly check-in,” in which they present their group’s work over the past week to the rest of the domain. To prepare for the check-in, each group must submit a weekly Participation assignment 24 hours before discussion section.
The weekly check-in gives you an opportunity to practice talking about technical material, solicit feedback from your mentor, and learn from other teams’ successes and obstacles. As such, attendance in discussion each week is mandatory, like in Quarter 1. If you’re unable to attend in a given week, work out an arrangement with your mentor in advance.
See the Winter 2023 tab of the Domains and Mentors spreadsheet for the time and location of your discussion meetings, as they may have changed since the fall.
In addition to any office hours your mentor may hold, methodology course staff will also hold office hours to help you resolve technical problems with your deliverables. See the Office Hours page for the schedule, and the Staff page for a listing of each TA’s area of expertise. Note that each domain will have an assigned methodology TA starting Week 2, who you will be required to check in with periodically during the quarter.
Note: These weights may change slightly during the first two weeks of the quarter.
|Each component of the Quarter 2 Project has a checkpoint to ensure that you’re on track.
|If your project is a traditional methods or analysis project, your primary deliverable is your report and your secondary deliverable is your website.
If your project is building a product (e.g. an application or dashboard), your primary deliverable is your product itself and your secondary deliverable is your report, though you will still have to create a separate website. See the Quarter 2 Project spec for more details.
|Your produced poster and presentations at the in-person capstone showcase will be graded on polish and clarity.
|Your final code submission will be graded based on the methodological standards introduced in Lessons 2, 5, and 6 of Quarter 1.
|Weekly prompts that require you to document each group member’s progress so far and how your project plans have evolved.
|As in Quarter 1, your mentor will assign you an overall participation grade at the end of the quarter.
Unless otherwise stated, we will follow the same General Rubric as in Quarter 1. We’ll also follow the same policy on Letter Grades. Note: your grade in DSC 180B is completely independent of your grade in DSC 180A.
Refer to the Collaboration Policy and Academic Integrity from Quarter 1.
Before the quarter begins:
- Confirm the date and time of your discussion section in the Domains and Mentors spreadsheet, as it may have been updated since you last checked. g
- Note that sections begin the week of September 26th (Week 1).
- Some sections have merged, including B02/B05 and B03/B19.
- If your section is held in SDSC, follow the instructions found here to enter.
- Note that the first lecture will be held live in-person on Monday, September 26th.
- Make sure you can access the following sites; email Suraj if you can’t:
Note that we will not be using Canvas at all this quarter.
As mentioned in the [Overview] overview), the primary goal of Quarter 1 is to get you acquainted with your mentor’s domain. The main deliverable in Quarter 1 is the “Quarter 1 Project”, which you will work on gradually throughout the quarter by completing the tasks that your mentor assigns you. The Quarter 1 Project is due at the end of the quarter, but a checkpoint is due in Week 5, to make sure you’re on track.
You will complete your Quarter 1 Project either individually or in groups, depending on your mentor’s preferences. Your Quarter 1 Project will serve as a foundation for your Quarter 2 Project Proposal, which you will submit in Week 9. The Quarter 2 Project will be completed in groups of 2-4 throughout Quarter 2.
Note: You may not get to choose who your “partners” are, as that may be up to your mentor; like in industry or academia, groups will be formed using a variety of factors, including academic background, mutual interests, and perhaps a little randomness.
The subsections below describe how the course operates.
Lectures are focused on methodological skills that can apply to all domains. In lectures, we will cover best practices with software engineering for data science and project management (see the course homepage for a full listing in Quarter 1).
Based on feedback from prior iterations of the capstone, we’ve decided to deliver methodology lectures as lecture notes that you read outside of class and discuss with course staff during synchronous sessions. Specifically:
- Each Sunday, we will post a “lesson” on the course homepage. A lesson will contain all of the methodology content you need to learn for the week. Read each lesson on your own. (We will make an announcement on EdStem when we release new lessons).
- The day after, on Monday, you may attend optional synchronous “Lecture Help” sessions during the scheduled lecture slots. These are scheduled on Mondays from 3-3:50PM and 4-4:50PM, both in Mandeville Center B-210 (map). In these sessions, we will answer any questions you have with the lesson.
- You can attend either (or both) of the sessions, regardless of which lecture section you’re enrolled in.
- Note that in Week 1 (on September 26th), we will hold a “traditional” introductory lecture during the scheduled lecture slots, and you should plan to attend. The plan mentioned above begins with Week 2.
- All lecture sessions will be podcasted, but the usefulness of the podcasts for Lecture Help sessions will be limited, since they will primarily contain Q&A amongst the students present.
- Many methodology lectures will have an accompanying “methodology assignment”, due the following weekend. See the [Assignments] assignments) section for more details.
- In the Lecture Help sessions, we will provide help on methodology assignments as well – expect the assignments to be much easier to complete if you come to Lecture Help sessions!
Note that we will not be using the “lab” component of the course that appears on WebReg.
Each week, you will meet with your domain mentor for an hour in discussion section. You can see the time and location of your discussion meeting in the Domains and Mentors sheet. Unlike Lecture Help sessions, attendance in discussion section is mandatory, and you must notify your mentor in advance if you can’t make it in a particular week. (If you have a permanent time conflict with your discussion section, you should switch to another domain.)
Each week, your domain mentor will assign you a combination of readings and tasks to complete, along with “participation” questions to answer to ensure that you’ve engaged with the material. You must complete these participation questions 24 hours before discussion, as your responses to them will drive the class discussion. Your mentor may provide you with specific participation questions to answer; if not, you should answer these “default” participation questions. You should complete the other tasks they assign you before discussion as well, though you may not have to submit them anywhere. Later in the quarter, you will brainstorm project proposals in discussion as well.
Note that discussion section will consist of discussion, not lecture. As such, if you do not ask questions in discussion section, no discussion will occur. To get the most out of the capstone program, you should actively participate in discussion section. In the workplace, you will often need to communicate with your coworkers and ask questions when you don’t understand things, and the same is true here.
There are two flavors of office hours:
- Methodology office hours, held by the methodology (DSC 180A) course staff. Come to these office hours with questions on methodology lectures and assignments, and on how to apply methodology concepts to your domain work. See the Office Hours page on the course website for the schedule.
- Note that different TAs have expertise in different areas; see the Staff page for a listing on each TA’s area of expertise.
- Domain office hours, held by your domain mentor. Come to these office hours with questions on the readings or tasks your mentor assigned you or on your projects. See the Domains and Mentors sheet or your section’s website (if applicable) for the schedule.
- You will be required to attend your mentor’s office hours at least three times throughout the quarter.
As is common in data science, you will likely find yourself as a bridge between domain specialists and (computing) methodology specialists. In this course, it is totally normal if your domain mentors do not know specifics of your code (or even know the language you are coding in!). You will have access to help from both methodology experts (in office hours and EdStem) and your domain mentor (in discussion section and office hours). As such, it is up to you to formulate your questions for the appropriate audience (methodology expert or domain expert) so that you can adequately communicate with them to solve the problems you are facing.
The table below summarizes all the ways you will be assessed in Quarter 1.
|Assignments that develop your software development and project management skills. Submitted individually.
|Participation questions (default questions or mentor-provided)
|Weekly questions to keep you engaged with the material and to inform your mentor of class progress. Submitted individually.
|Weekly, 24 hours before discussion
|Graded for completeness by methodology TAs
|Engaging in conversation in discussion section is important for success in the capstone; as such, your mentor will assign you an overall participation grade at the end of the quarter.
|Graded by mentors
|Quarter 1 Project
|A chance to put together everything you’ve produced while learning about your domain. Submitted individually or in groups (up to mentor).
|Week 5 (checkpoint), Week 10 (final submission)
|50%: reports, graded by mentors (checkpoint + final)
20%: code, graded by methodology TAs (checkpoint + final; graded to ensure best practices are followed)
|Quarter 2 Project Proposal
|Proposal for final capstone project. Submitted in groups.
|Graded by both methodology TAs (elevator pitch) and domain mentors (schedule + write-up)
Note that the table contains links to assignment descriptions; the Quarter 1 Project and Quarter 2 Project Proposal details are tentative, and won’t be finalized until they are officially released. We will make EdStem announcements when these components are finalized.
In order to ensure consistent grading across such a diverse array of domains, we will utilize a coarse grading scheme with a clear rubric. This scheme will reflect broad checkpoints that you meet, and should help maintain focus on large, impactful things that you can improve on while reducing grading disagreements.
The grading scheme we will use for all assignments (other than for participation questions, which are pass/fail, and methodology assignments, which may have numerical scores) follows an A/B/C/F scale (without plus/minus), developed by Shannon Ellis:
|Accomplishes the task accurately, completely, and clearly. Code is clear, effective, and efficient. Written component is concise, at the appropriate level, and correct. Oral component (when applicable) is effective and within the time limit.
|Accomplishes the task well, but lacks some completeness or clarity. Code runs but lacks some aspect of clarity, effectiveness, and or efficiency. Written component is logical and generally correct, but lacks either clarity or accuracy. Oral component (when applicable) is moderately effective and/or slightly outside the time window.
|The task is somewhat accomplished, but lacks significantly when it comes to completeness and clarity. Code present but does not accomplish the task up to the standards of a data science graduating senior. Written component lacks substantial clarity/correctness. Oral component (when applicable) significantly lacks effectiveness/clarity.
|The task largely remains unaccomplished. Code lacks completeness, structure, and is unclear. Written component lacks required information to understand the work done. Oral component (when applicable) is nonsensical/unclear.
Individual assignments will be graded on the A/B/C/F scale above, and your overall course grade will be determined by using the proportions listed at the start of this section. For the purposes of computing your course grade, A, B, C, and F map to 4, 3, 2, and 0. So for instance, if you earn:
- full credit (A) on methodology assignments,
- an A on participation questions,
- a B on participation (as graded by your mentor),
- an A on your Quarter 1 Project, and
- a B on your Quarter 2 Project Proposal,
your “numerical” grade would be \(0.05 \cdot 4 + 0.05 \cdot 4 + 0.05 \cdot 3 + 0.7 \cdot 4 + 0.15 \cdot 3 = 3.8\).
You are guaranteed to earn at least the letter grade that your numerical grade converts to. For instance, a 3.7 is guaranteed to learn at least an A-, and a 2.0 is guaranteed to earn at least a C. When your numerical grade is between two letter grades, you are guaranteed to earn at least the lower letter grade; for instance, 3.8 is between 3.7 (A-) and 4.0 (A), so a 3.8 is guaranteed to learn at least an A-.
Note that at the end of Fall 2022 you will receive a grade in DSC 180A, and at the end of Winter 2023 you will receive a grade in DSC 180B; these are two separate courses, each worth 4 units.
With all of that said, in this course, you should not worry about your letter grade. The grades you receive on individual components of the course are meant to provide you with feedback on how to improve future submissions. To be successful in this course, you should strive to have engaging interactions with your domain mentor and to produce work that you are proud of. Nobody will remember whether you got an A- or a B in the capstone, but they will remember if you produce a stellar final project.
In DSC 180, we expect you to work hard and engage with material that originates outside the academic walls. All ideas and work must be your own, that of your approved group, or properly cited. Act with integrity and don’t cheat.
In DSC 180 you are encouraged to use outside resources to help with your work. However, you must properly cite any concepts, writing, or code that originates from other sources. If you are unsure of whether something needs a citation, it’s best to:
- consult the domain expert for your section,
- follow the examples in course readings, and
- place citations with relevant links in comments.
The following activities are considered cheating and ARE NOT ALLOWED in DSC 180 (this is not an exhaustive list):
- Using or submitting either writing or code acquired from other students (except your group, where allowed).
- Not properly citing ideas, writing, or code acquired from outside sources. (Citations are a good thing!)
- Having any other student complete any part of an assignment on your behalf.
- Completing an assignment on behalf of someone else.
The following activities are examples of appropriate collaboration and ARE ALLOWED in DSC 180:
- Discussing the general approach to understanding or solving a problem.
- Talking about debugging/cleaning strategies or issues you ran into and how you solved them.
- Using outside material with proper citations (including StackOverflow code!).