I am doing data analysis with jupyter notebook using python and pandas. In my master csv, thousands are indicated with dots and python reads the columns as a float. I want t