Branch-AI/one_exercise_per_file/week01/day04/ex04/audit/readme.md

The for loop is forbidden in this exercise. The goal is to use groupby and apply.

1. This question is validated if the output is:

       df = pd.DataFrame(range(1,11), columns=['sequence'])
       print(winsorize(df, [0.20, 0.80]).to_markdown())
   

   	sequence
   0	2.8
   1	2.8
   2	3
   3	4
   4	5
   5	6
   6	7
   7	8
   8	8.2
   9	8.2

2. This question is validated if the output is the same as the one returned by:

   def winsorize(df_series, quantiles):
   """
       df: pd.DataFrame or pd.Series
       quantiles: list [0.05, 0.95]
   
   """
   min_value = np.quantile(df_series, quantiles[0])
   max_value = np.quantile(df_series, quantiles[1])
   
   return df_series.clip(lower = min_value, upper = max_value)
   
   
   df.groupby("group")[['sequence']].apply(winsorize, [0.05,0.95])
   

   The output can also be a Series instead of a DataFrame.

   The expected output (first rows) is:

   	sequence
   0	1.45
   1	2
   2	3
   3	4
   4	5
   5	6
   6	7
   7	8
   8	9
   9	9.55
   10	11.45

• https://towardsdatascience.com/how-to-use-the-split-apply-combine-strategy-in-pandas-groupby-29e0eb44b62e