Final Peer feedbacks - Githubissues

UBC-MDS / DSCI_522_group_401

Medical Expenses Prediction

MIT License

0 stars 4 forks source link

Final Peer feedbacks #30

Closed sirine-chahma closed 4 years ago

sirine-chahma commented 4 years ago

feature importance plot (lollipop chart for example) (Sree)
check the correlation between features (for example correlation heatmap,...) (Sirine)
remove the BMI VS age plot (Sirine)
split the BMI in 3 categories and treate it as a categorical variable (Sirine)
do a one side hypothesis test for the smoker VS non-smoker test (Karan)
add the line y=x on the actual VS predicted expenses (Sree)

sirine-chahma commented 4 years ago

The correlation heatmap is done

sirine-chahma commented 4 years ago

I removed the BMI VS age plot

sirine-chahma commented 4 years ago

I split the BMI in 4 categories and updated the corresponding charts

singh-karanpal commented 4 years ago

I have updated following to the project:

Makefile workflow
One Side Hypothesis Test
Added Test Case to EDA

singh-karanpal commented 4 years ago

We have addressed all of the above concerns in our Milestone 3. At this point in time, we have addressed TA feedbacks and Peer feedbacks too. We can close this issue now.