Project 1 Summary
Overall the class did an excellent job!
21/29 were 19 or higher!
19+ (21 projects): 20, 20, 20, 20, 20, 20, 19.75, 19.5, 19.5, 19.5, 19.5, 19.5, 19.5, 19.5, 19.5, 19.5, 19.5, 19.25, 19.25, 19, 19
18+ (5 projects): 18.75, 18.75, 18.75, 18.5, 18
General Comments: Kudos and Cautions
The scores were very good, and, generally speaking, people are understanding the material, we want to mention a few issues that came up.
Kudos
Several people used ordinal encoding on the columns instead of one-hot encoding. One person even experimented with both ordinal and one-hot encoding to see which one performed best. Excellent job thinking about the data and what might make the most sense for
Several people experimented with other techniques, such as other kinds of ML models. This was not required but we are always happy to see students eager to experiment with new kinds of techniques. A word of caution though: be sure you understand the API you are using, and feel free to talk to us if you’re interested in exploring some new API or techniques.
Cautions
When using AI, make sure you understand what the generated code does. It’s clear several people are using ChatGPT, etc., but some of you seemed to Also, please state exactly how you used it. Do this by adding comments directly in the code (i.e., Jupyter notebook) describing what function/code blocks were generated, etc., by AI. We did not count off on Project 1 for this but we will be in future projects.
Many people did not realize that they needed to treat columns with invalid values (“?” and “*”) – you can use the
unique()function on a dataframe to see all values.Please consider splitting code into multiple cells and adding comments to the code. This will improve the code readability.