1  import pandas as pd
 2  from sklearn.impute import KNNImputer
 3  
 4  dataframe = pd.read_csv("credit_risk_dataset.csv")
 5  
 6  # Feature Conversion Values
 7  person_home_ownership_values = {
 8      "RENT": 1,
 9      "MORTGAGE": 2,
10      "OWN": 3,
11      "OTHER": 4,
12  }
13  loan_intent_values = {
14      "EDUCATIONAL": 1,
15      "MEDICAL": 2,
16      "VENTURE": 3,
17      "PERSONAL": 4,
18      "DEBTCONSOLIDATION": 5
19      }
20  loan_grade_values = {
21      "A": 1,
22      "B": 2,
23      "C": 3,
24      "D": 4,
25      "E": 5
26  }
27  cb_person_default_on_file_values = {
28      "Y": 1,
29      "N": 0,
30  }
31  
32  dataframe["person_home_ownership"] = dataframe["person_home_ownership"].map(person_home_ownership_values)
33  dataframe["loan_intent"] = dataframe["loan_intent"].map(loan_intent_values)
34  dataframe["loan_grade"] = dataframe["loan_grade"].map(loan_grade_values)
35  dataframe["cb_person_default_on_file"] = dataframe["cb_person_default_on_file"].map(cb_person_default_on_file_values)
36  
37  imputer = KNNImputer(n_neighbors=9, weights="uniform", metric="nan_euclidean")
38  imputed_data = imputer.fit_transform(dataframe)
39  pd.DataFrame(imputed_data,
40               columns=dataframe.columns).to_csv("imputed_data.csv", index=False)