PANDAS ELIMINATING TEMP2, TEMP2, TEMP1

cities[(cities.latitude > 50) & (cities.temperature > 9)]\
[['city','longitude']]\
.sort_values('longitude')

PANDAS YOUR TURN #1

f = open('Countries.csv')
countries = pd.read_csv(f)
countries[(countries.EU == 'no') & (countries.coastline == 'no')][['country','population']].sort_values('population')

PANDAS YOUR TURN #2

countries.groupby('coastline').mean().population
countries.groupby(['coastline','EU']).mean()[['population']]

PANDAS YOUR TURN #3

citiesext = cities.merge(countries, on='country')
print('EU before Brexit:', np.average(citiesext[citiesext.EU == 'yes'].temperature))
print('Non-EU before Brexit:', np.average(citiesext[citiesext.EU == 'no'].temperature))
print('EU after Brexit:',\
   np.average(citiesext[(citiesext.EU == 'yes') & (citiesext.country != 'United Kingdom')].temperature))
print('Non-EU after Brexit:',\
   np.average(citiesext[(citiesext.EU == 'no') | (citiesext.country == 'United Kingdom')].temperature))

PANDAS YOUR TURN #4

Kcities = cities[cities.city.str.contains('k')].sort_values('temperature')
print('Second-coldest:\n', Kcities.head(2).tail(1)[['city','country']])
print('Second-warmest:\n', Kcities.tail(2).head(1)[['city','country']])