Unit Project
For all of the BAH DC cohorts: Wednesday, June 6th by midnight.
Submissions must be made via a push to this repository on your own fork. This time, late submissions and submissions via Slack and email will NOT! be accepted, no exceptions!
| Topic | Description | Link |
|---|---|---|
| Starter Code | Project Description | Here |
| Solution Code | Sample solutions for project questions | Here |
| Dataset | Boston Housing Dataset (Provided as back-up to embedded data) | Here |
Note: Instructors should withold providing project solutions until students have submitted their drafts. In order to do so, avoid forking this repo and instead clone down materials locally, create a new repository in your student org, and upload materials manually.
In this project, you'll build a linear regression (and KNN) model using guided prompts. You will be exposed to model validation using the train/test split (and K-Fold methods as a bonus) in order to fit different models on sample Boston housing data.
This project includes required and bonus objectives. You only need to complete the required portions, but you're encouraged to try as many of the bonus questions as possible!
For this project, you will be:
- Creating a train/test split of the data.
- Training each of your models on the training data.
- Evaluating each of the models on the test data.
- Ranking models by how well they scored on the testing data set.
In a Jupyter Notebook, create working solutions for all of the required questions. Your notebook should include:
-
Text for each question, copy and pasted from the starter code provided.
-
A working solution to each problem.
- Do not include test, practice, or broken code (unless you were unable to create a working solution).
-
Comments for all of your code.
- In your comments, describe any assumptions you made in order to solve these problems.
-
Bonus: After completing the required portions, try your hand at the bonus sections for some additional challenges!
For all projects, requirements will be evaluated on a simple point scale of 0, 1, or 2. Additionally, instructors will provide you with feedback on required portions of your project.
| Score | Expectations |
|---|---|
| 0 | Incomplete. |
| 1 | Does not meet expectations. |
| 2 | Meets expectations, good job! |
| 3 | Surpasses our wildest expectations! |
Note: Scores of
2mean that a requirement has been completely fulfilled, while3is typically reserved for bonus objectives.
Your instructor will explain how to submit your assignment. Typically, this is done either by:
- Creating a repository in your github profile, hosting your materials, and sharing a link with your instructor. [or]
- Forking the project repository, adding your solutions, and submitting a pull request back to the relevant repo.
- Make sure to provide interpretation for all outputs.