I've used skipping the line to avoid the issue of reading the data.

df = pd.read_csv("file_name.csv",on_bad_lines='skip') 

However, I believe it depends on the data we are working on.