search

UBC-MDS / DSCI_522_Group-308_Used-Cars

This project attempts to build a regression model to predict price of used cars based on numerous features of the car
MIT License
2 stars 6 forks source link

Milestone 3: Report #57

Open ksedivyhaley opened 4 years ago

ksedivyhaley commented 4 years ago

Reasoning:

You've done a good job of addressing my feedback. I'd like to see you address more peer feedback on this area. You've closed issues #40, #41, and #43, but I'm not seeing these issues being addressed in your report. If they're not feasible to implement, mention and briefly discuss under further directions (and leave the issues open).

"Eg., having an RMSE almost twice higher than MAE suggests that there is a good number of observations where the error is big (the more RMSE differs from MAE, the higher is the variance)"

I would still like to see values given for RMSE and MAE

Writing: Much improved but still minor typos

bradentam commented 4 years ago

We actually did implement these some of these issues that were more pressing. I closed them because most of the issues had to do with further improvements and adding features to the project which we cannot address due to time constraints. I was planning on formalizing a "further directions" issue after milestone 4.

For milestone 3 report we:

ksedivyhaley commented 4 years ago

Hi Braden, thanks for the clarification, I see those changes now.

Where the other feedback is concerned, it is generally good practice not to close issues that you plan to address later. I expect from the point of view of milestone task management it's nice not to have them clutter up your open issues - perhaps a label would be useful here to differentiate between things that need to be done for the next milestone and things that don't?

With respect to the RMSE and MAE, the table of values is a good inclusion. As a minor point, with respect to the sentence in your conclusion:

Although we achieved a solid R^2 value of 0.816, we can now observe some other metrics. Eg., having an RMSE almost twice higher than MAE

It would be nice to repeat the numeic values for RMSE and MAE in this sentence, as you repeat the R^2 value.