[Python] Pandas read_csv: Skip rows contains invalid data that can cause data_type parsing...

The csv file can contain string values to certain integer columns and i want to ignore/handle via callback if that happens, tried using on_bad_lines='skip/warn' however it gets triggered only on wherever there are parsing issues due to delimiter/num of columns mismatch. Here is the code i am using

csv:

id,name,age
1,,"25"
2,"",30
jj,John,



import pandas as pd
schema = {'id':'int64','name':'str','age':'int'}
df = pd.read_csv('a.csv', dtype=schema, on_bad_lines='warn', engine='python')

Continue reading...