Pitches Solution

Solution Files

The following files represent an example solution for the Pitches lab.

Note that many deviations from the solution presented here are acceptable, and you should be cautious of overfitting to these documents when preparing future reports.

The solution presented will also likely contain more depth and detail than is expected of students. The solution is an example of a report quality that students should aspire to, but are not yet expected to be able to produce.

Instructor’s Notes

Reports must be rendered via Quarto. We saw many reports that were instead rendered with the default Jupyter renderer.
Be sure to keep the YAML header provided by the template notebook!
- Alternatively, if you are not using the template, but sure to add the YAML header as described in the lab policy document.
- Failure to do so will result in an ill-formatted and include rendered report, that will result in significant point loss.
Be sure to submit both the requested notebook (.ipynb) and rendered report (.html) files.
When writing your report, you should write as if the prompt for the lab does not exist. Perhaps more clearly: you should go a step further and write as if CS 307 did not exist. Recall that each lab presents a scenario and goal that includes a description of who you are, and who you are writing for.
- We saw many titles that were simply “Lab 01”. While that is a title in the sense that there was text present in the main heading, it does not satisfy the need for a title.
- You should not in any way mention the production data.
- You should not reference the performance metrics given in the lab prompt. In practice, it is unlikely that you’ll be given a task with specified performance metrics. Instead, you’ll often need to determine if a model performs good enough given the context of the problem it is solving, which will require the analyst to determine what metrics are acceptable. You will be asked to do exactly that for the final project.
While use of AI is currently not forbidden in CS 307, we detected a number of reports that show signs of what we feel is inappropriate AI usage. We do not want to spend time determining if work was created via AI, but if necessary, we will consider altering course policies to put specific limitations on AI usage. In general, we hope that AI is used to assist you with your work, but not simply do your work for you.
The Quarto documentation on Execution Options details methods for controlling how code cells are executed and displayed. If code is emitting warnings or messages you would like to suppress, the easiest method is to use #| output: false.
While the lab policy document suggests specific filenames, note that using alternative files will not impact grading. Unfortunately, Canvas does not provide students with full control over filenames, so we cannot set any filename requirements.
In general, the discussion section should be the most thorough (and thus “biggest”) of those you are asked to write.
You might notice some additional code in the YAML header of the solution notebook. Specifically we added enabled: true under the execute: key. This has two effects:
- It allows for inline code to be used.
- It re-runs all cells when rendering or previewing the document.
  - This has the benefit of insuring that your cells were run in order, but costs the time to re-run those cells!
In the solution notebook, you will notice that in addition to figures being captioned via Quarto with fig-cap, we also give those chunks a label. This allows us to reference figures in the text with expressions like @fig-velocity-spin. Similarly, we build tables and caption them with tbl-cap and use labels for referencing them in the text.