'nan' in GDP

[evanp]

This doesn't seem to be a valid actor_id. Is this "not a number"?

[lgloege]

Two countries were the problem, Kosovo and Namibia.

• ISO code for Namibia is NA, which was being parsed as NaN. The solution there is to set keep_default_na to False. df= pd.read_csv('file.csv', keep_default_na=False) [Easy fix]
• Kosovo is not in our ISO-3166 database, we should add this (https://laendercode.net/en/2-letter-code/xk#:~:text=Country%20Code%20XK%20Country%20code%20according%20to%20ISO%2D3166%20Alpha%2D2&text=XK%20is%20the%20two%2Dletter%20country%20abbreviation%20for%20Kosovo.). I created a ticket for this OC-1024

However, adding Kosovo is a low priority since i'm not seeing it in our emission or pledge datasets