Your performance will be evaluated according to your predictions on the test data. We will provide the entire test dataset but with the target column omitted. You must train your machine learning model on the training set you have been working on to predict the targets on the test data and submit your predictions.

## Grading
As mentioned on [Canvas](https://wustl.instructure.com/courses/97407/assignments/440384?module_item_id=1438094), Milestone 3 is worth 20 points, consisting of:

- 8 points for code/kernel and report
- 6 points for beating baseline 1
- 6 points for beating baseline 2
- 2 (bonus) points for first, second, and third place submissions

The 8 points for code/kernel and report rely mostly on your written report, but a small part of it is the documentation you provide in your code.

## Metric

The evaluation metric for this competition is [RMSE](https://en.wikipedia.org/wiki/Root-mean-square_deviation). The RMSE, commonly used for regression tasks, is the square root of the mean-squared error.

## Baselines

Baseline 1: 0.2000
Baseline 2: 0.1700

## Public vs. Private scores

When you submit your predictions, you'll have a public score based on a subset of the test data (50%) and a private score based on the rest. You'll be able to see your public score as soon as you submit a prediction, but the private score will only be revealed after the deadline. The private score will determine your final performance in this competition.

## Submission format

Submission files should contain two columns: "Id" and "target", and the file should contain a header and have the following format:
```
Id,target
1,0.1
```
Note that due to Kaggle's own limitations, you are limited to 20 submissions each day (UTC). You may choose which submission to be evaluated. Otherwise, the system will simply choose the best one.

## Code/Notebook

To verify your work, you must create a notebook on Kaggle that generated your submission, or link your github in the report. If you're using Kaggle to submit your code, be sure to share it with the following username for grading: cynthiazma