How grading works in this course¶
Each graded assignment receives one final letter grade. Projects 1–5 use four broad letter-graded dimensions; the weighted average becomes the project letter. Code Review Presentations, Growth Reflections, the proposal pitch, and the final project are holistically assessed as described below.
The scale¶
| A+ | A | A− | B+ | B | B− | C+ | C | C− | D+ | D | D− | F |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 4.3 | 4.0 | 3.7 | 3.3 | 3.0 | 2.7 | 2.3 | 2.0 | 1.7 | 1.3 | 1.0 | 0.7 | 0 |
Your grade in each component is the average of the letters you earned in it, and your course grade is the weighted average of those. Letters average the way you would expect: an A+ and an A− come out to an A. Grades are recorded as these numbers in Canvas, so you can see where you stand at any point.
A+ is not “did extra.” It is for work showing judgment beyond what was asked — a validation you thought of yourself, a failure mode you found and explained, a simpler approach than the one specified. Volume does not earn it.
Turning your average into a course grade¶
Each cutoff sits 0.2 below that grade’s own value, so you can fall a little short of straight-A work and still earn an A. What it will not do is round you up a whole step: consistent A− work earns an A−.
| Average | Grade | Average | Grade | |
|---|---|---|---|---|
| ≥ 3.80 | A | 1.80 – 2.09 | C | |
| 3.50 – 3.79 | A− | 1.50 – 1.79 | C− | |
| 3.10 – 3.49 | B+ | 1.10 – 1.49 | D+ | |
| 2.80 – 3.09 | B | 0.80 – 1.09 | D | |
| 2.50 – 2.79 | B− | 0.50 – 0.79 | D− | |
| 2.10 – 2.49 | C+ | < 0.50 | F |
Weights¶
| Component | Weight |
|---|---|
| Projects 1–5 (8% each) | 40% |
| Code Review Presentations (5% each) | 10% |
| Final project (proposal pitch 5%; finished project 25%) | 30% |
| Growth Reflections (3) | 10% |
| Scholarly engagement | 10% |
What the grades mean¶
The descriptions below are anchors. Plus and minus grades indicate work between adjacent descriptions; A+ is reserved for the judgment described above.
Projects 1–5¶
Project grade calculation¶
The canonical project dimensions and weights live on the Projects page. For each dimension, I assign a letter using the course scale above, then calculate:
where is correct model and validation, scientific reasoning, reproducibility and code design, and communication and figures. I translate the resulting score to a final letter with the cutoff table above.
For example, an A in model/validation, B+ in scientific reasoning, A− in reproducibility, and A in communication produces , which is an A−.
Letter anchors for project dimensions¶
Use these anchors for all four dimensions. Plus and minus grades represent work between adjacent anchors; A+ requires judgment beyond what was asked.
| A | It works, you showed that it works, and someone else could pick it up and run it. The validation is real — convergence at the expected order, a limiting case checked, a comparison against something that could have disagreed. The written interpretation explains what you did and what it means. |
| B | It works and you checked it, but the evidence is thinner than the claim, or the code takes effort to follow, or the writing leaves me guessing about a choice you made. |
| C | It runs, but there is little evidence it is right — tests that would pass for a broken implementation, or results reported without any check that could have caught an error. Or pieces are missing. |
| D | Substantially incomplete, not working, or missing what was asked for. |
| F | Not submitted, or not your own work. |
Each project’s analysis notebook includes written reasoning and interpretation in the places required by the project brief. Those answers are evidence for the scientific-reasoning and communication dimensions; they are not a separate points category.
Code Review Presentations¶
Each of the two individual presentations receives one holistic letter grade and counts for 5% of the course grade. I look for a useful map of the package, a specific code trace, a check that could have disagreed with the student’s expectation, and a conclusion that stays within the evidence. See the assessment brief for the rounds, format, and grading anchors.
Final project¶
The final project is still judged holistically: does the code do the right thing, do you know that it does, and could someone else follow it? Its brief will define its evidence and communication expectations.
Final-project proposal pitch¶
The five-minute in-class proposal pitch is judged as a whole: is the question clear, the scope feasible, and the plan capable of producing evidence? Slides are a communication aid, not a separate written assignment.
Growth Reflections¶
Growth Memos 1 and 2 should each be 1–2 pages, with a minimum of one full page. The Growth Synthesis should be 2–3 pages, with a minimum of two full pages. They are judged on specificity, reflection, and evidence, not literary polish. Submit each through the corresponding Canvas assignment as a single-spaced PDF using a readable 12-point font and 1-inch margins.
| A | Specific and reflective. Uses concrete evidence to explain how you worked, what you learned, and what changed or stayed the same. |
| B | Relevant but general. Makes reasonable claims without enough specific evidence or reflection to show what they mean for your work. |
| C | Minimal or largely descriptive, with little specific evidence or reflection. |
| D | Substantially incomplete or too vague to demonstrate meaningful reflection. |
| F | Not submitted. |
“I did not understand why Newton’s method diverged until I plotted the iterates” is worth more than a page of general statements about learning a lot.
| Assigned | Due | |
|---|---|---|
| Growth Memo 1 — baseline | Tue Aug 25 | Mon Aug 31, 11:59 pm PT |
| Growth Memo 2 — mid-semester | Tue Oct 6 | Mon Oct 19, 11:59 pm PT |
| Growth Synthesis | Tue Nov 17 | Mon Dec 14, 11:59 pm PT |
Scholarly engagement¶
One grade, at the end of the semester, for the term as a whole.
| A | Consistently prepared. Contributes through questions, visible reasoning, productive collaboration, or substantive check-ins. |
| B | Steady and cooperative, with occasional unpreparedness or limited visible engagement. |
| C | Frequently unprepared or disengaged; contributes mainly when prompted. |
| D | Rarely prepared or engaged, with little visible participation in the work of the course. |
| F | No meaningful evidence of scholarly engagement. |
Attendance itself is not graded. However, excessive absences will affect Scholarly Engagement because they limit prepared participation, collaboration, substantive check-ins, and other visible engagement with the course.
Submission¶
Project 1 is submitted by linking the GitHub repository in its Canvas assignment. Later project briefs will name their submission route, including any Classroom50 use. Code Review Presentations and Growth Reflections are submitted through their corresponding Canvas assignments. The final-project brief will specify its submission route. Projects are typically due Wednesdays at 11:59 pm PT. The Schedule lists each specific deadline.
For projects, your submission is the latest commit pushed to GitHub by the deadline. Work that remains only on your computer has not been submitted. Push early and often: a pushed work-in-progress always beats a perfect uncommitted local copy.
See the syllabus for extensions and regrade requests.