Portfolio Project: U.s Medical Insurance: Peer Review please!

Projects Portfolio Project: U.S. Medical Insurance Costs
1

Hello Everyone! I would really appreciate a code review of my us medical project. Any criticism is welcome. Thank you!!

2

Congrats on completing the project.

Some thoughts-

  • Rather than jumping right into the functions, it might be a good idea to run some basic descriptive stats on the data first (read up on EDA-exploratory data analysis). That way, anyone viewing the project can get a general idea of exactly what’s in the data, then you can go about analyzing differences in the two populations you selected from the data.
    • Stuff like, how many smokers vs. non-smokers are in the data? How many men vs. women? What’s the median charge and how is that different from the mean? The mean is pretty high, so that might be an indication of outliers in the data, which would pull the mean.

All of these built in methods are available in Pandas.
https://pandas.pydata.org/docs/user_guide/index.html#user-guide

Ex:

df.describe()
>>	     age	    bmi	       children	   charges
count	1338.000000	1338.000000	1338.000000	1338.000000
mean	39.207025	30.663397	1.094918	13270.422265
std	14.049960	6.098187	1.205493	12110.011237
min	18.000000	15.960000	0.000000	1121.873900
25%	27.000000	26.296250	0.000000	4740.287150
50%	39.000000	30.400000	1.000000	9382.033000
75%	51.000000	34.693750	2.000000	16639.912515
max	64.000000	53.130000	5.000000	63770.428010

or,

df['age'].mean()
>39.2

df['charges'].mean().round(2)
>>13270.42

df['charges'].median().round(2)
>> 9382.03


df["sex"].value_counts()
>male      676
female    662

df["smoker"].value_counts()
>>no     1064
yes     274

etc. etc.
  • Use text cells to lay out your thought process for the reader while you comb through the data. You’re ultimately telling a story here with the analysis.

Also, just a suggestion for the readme file–you don’t need to mention how long the project took you as that’s irrelevant to anyone viewing it. I would also avoid mentioning CGPT, as who knows the accuracy of the results, or what’s actually in the training data in the model.

1 Like
3

Thank you @lisalisaj I am looking into EDA I see what you mean! Hopefully I will get a better understanding of it and use it properly to evaluate my code. I look forward to another review from you. Thank you for your help!!

1 Like