Update values of a dataframe based on conditions from another dataframe

Question

I'm trying to map values (Weight of Evidence) from a dataset to another.

My original dataset looks as follows:

dataframe 1

The aim is to substitute these rows with the corresponding weight of evidence values.

I would do it manually, for example:

df.loc[df['loan_type'] == 1, 'loan_type'] = 0.008241

I'm looking for a better way to do it (not manually) Similar to this post: Python Dataframe: Update the values of a column in a dataframe based on another dataframe

Dataframe 2

Answer 1

This should do the trick:

df = pd.DataFrame({'loan_type':[1,2,1,3], 'var_1':[1,0,1,2], 
               'var_2':[1,1,1,1],
               'var_3':[1,1,1,1],
               'var_4':[1,0,1,0],
               'target':[0, 0, 0,0 ]})

woe_df = pd.DataFrame({'variable':['loan_type', 'loan_type', 'loan_type', 'var_1', 'var_1', 'var_1', 'var_2', 'var_3', 'var_4', 'var_4'],
               'modality':[1,2,3, 0,1,2, 1, 1, 0, 1], 
                'woe':[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1]})

woe_mapping = defaultdict(dict)
for variable in woe_df.variable.unique():
    var_woes = woe_df[woe_df.variable==variable]
    for modality, woe in zip(var_woes.modality, var_woes.woe):
        woe_mapping[variable][modality] = woe
    else:
        df[variable] = df[variable].map(woe_mapping[variable])