Milestone 2 Report Feedback

[ahofmann4]

Hi Group 16!

I have a lot of suggestions for you, and part of the confusion on my end is the fact that your 'Report' document is more like an eda report, but with some report sections. By making a separate Report document and putting it into the doc folder much of the confusion will be fixed. Your new Report file should be linked to your ReadMe file

• report missing. URL in ReadMe leads to a file that is more of an overview of the EDA, there are some components of the report, but overall a lack of explanation of EDA findings, and exploration/justification of the use of all features.
• Table of Contents Missing, Names of Contributors missing, data of last change missing,, figure legends/titles missing, I'm not sure all the distribution plots of the features are useful to me to look at in the report, and I'm wanting to know about how some of the features correlate with each other
• eda file jumps from nice explanation of the data structure, into eda figures of unknown significance, to a short walk through of the model fitting
• The figures showing the distributions, I have to scroll back up to the top of the page to know what the 1 & 2 in the figure legends referred to, more explanation and text around these figures would be helpful final conclusion that the logistic regression is unclear as why it is the best model, Random Forest seems to be better in most columns

[IfyAnene7]

Hi @ahofmann4,

Thank you so much for the feedback! We will attempt to incorporate all these into the report before the final submission.

Regarding choosing Logistic regression as the best model, we explain why Random forest model was discarded (due to overfitting) in the last sentence of the first paragraph in the analysis and results section.