import pandas as pd
s_sep = pd.Series(['1,000,000', '1,000', '1'])
print(s_sep)
0 1,000,000 1 1,000 2 1 dtype: object
# print(s_sep.astype(int))
# ValueError: invalid literal for int() with base 10: '1,000,000'
print(s_sep.str.replace(',', '').astype(int))
0 1000000 1 1000 2 1 dtype: int64
print(s_sep.str.replace(',', '').astype(float))
0 1000000.0 1 1000.0 2 1.0 dtype: float64