Python : Sum dictionary values in a dataframe

I have a CSV file with the following layout /content
datetime,key_value_column
2022-02-02 00:00:01,"{‘key1′:2,’key2′:7,’key3’:100}"
2022-02-02 00:00:10,"{‘key5′:2,’key2′:3,’key3′:1,’key4’:5}"

The datetime column is unique

I would like to have python script that sums the values in the dictionary for each row ie
2022-02-02 00:00:01,107
2022-02-02 00:00:10,11

I can read the file

df = pd.read_csv('test.csv')
print(df)```

but no idea how to sum the dictionary values across the data frame  
I worked out summing values for a dictionary example below but I am  
 hoping to get some hints on how to process the dataframe  

```adict = {'key1':2,'key2':7,'key3':100}
sumvalues = adict.values()
print(sum(sumvalues))```

Thanks

>Solution :

If there are dictionaries and keys are numeric use lambda function only:

df['key_value_column'] = df['key_value_column'].apply(lambda x: sum(x.values()))

Or if there are string repr of dictonaries first replace them to dict by ast.literal_eval:

import ast

df['key_value_column'] = df['key_value_column'].apply(lambda x: sum(ast.literal_eval(x).values()))