解析数据框列以创建新的数据框

Question

I have a dataframe with colmuns that have nested attribute values pairs. How can I parse through this dataframe column and get these value to create a new dataframe for it?

This is one of the column values look like:

{'BikeParking': 'False', 'BusinessAcceptsCreditCards': 'True', 'BusinessParking': "{'garage': False, 'street': True, 'validated': False, 'lot': False, 'valet': False}", 'GoodForKids': 'True', 'HasTV': 'True', 'NoiseLevel': 'average', 'OutdoorSeating': 'False', 'RestaurantsAttire': 'casual', 'RestaurantsDelivery': 'False', 'RestaurantsGoodForGroups': 'True', 'RestaurantsPriceRange2': '2', 'RestaurantsReservations': 'True', 'RestaurantsTakeOut': 'True'}

Not all the columns have the same attributes as well. Please help, I am new in python.

Answer 1

If your dict is not nested, the following will work.

pd.DataFrame(list(df['column_name'])

If you have nested dicts, try doing the above again on the nested dicts.

Answer 2

It's not very clear what you're looking for but another thing that might help is using the apply function to parse every row of your dataframe to create a new column with the result.

def _parsing_function(row):
    // column containing nested values
    dict = row['column_name']
    for key in dict:
        'do something'
        return 'new column value'

frame['new_column'] = frame.apply(_parsing_function, axis=1)

Answer 3

To find the dictionaries inside your existing dataframe column and make new dataframes, you can iterate over your column and create dataframes as needed:

new_dfs = []   # list of new dataframes from dictionaries in existing column
for elem in df['column_name']:
    if type(elem) is dict:
        new_dfs.append(pd.DataFrame(elem))  # add to list of created dataframes