π₯ Week 05 - Competition
2022/23 Autumn Term
This week we are going to have a little competition. As part of π£οΈ Week 05 Lab, you will be asked to explore and propose a new logistic regression model that beats a baseline one we will provide.
The prize? A DSI water bottle for the person who achieves the highest F1-score in both training and testing sets using a legitimate logistic regression model.
Youβve got until Wednesday at 23:59 to submit your finest model via Moodle. I will reveal the winner in Fridayβs lecture.
The goal
Your mission now is to develop a model trained, as before, on data up to the end of 2018 that excels in F1-Score on the training and testing sets. Can we get closer to f_means=1?
Strategies to Consider
- Adjust the classification threshold
- Enrich the model by adding more variables.
- Transform variables β either manually or with
recipes::step_*functions.- For the recipes route, consult the sections on Individual Transformations and Normalisation on the
recipesdocumentation page. Just triple check the output of your recipes.
- For the recipes route, consult the sections on Individual Transformations and Normalisation on the
π Remember, the golden rule is you canβt use future data to forecast the past.