Complete the function below that takes in two frames and a list of column names. This function should return a new frame

that contains the left outer join between these frames on the specified columns from the LHS$.$

Hint: Make sure to read the documentation on DataFrame.join$()$ if you are going to use it$.$

$\text{'}\text{'}\text{'}$

def left$\_$join$($lhs$\_$frame, rhs$\_$frame, column$\_$names$)$:

return NotImplemented

print$("$World data that includes Covid$-19$ stats:"$)$

covid$\_$world$\_$data $=$ left$\_$join$($world$\_$data, who$\_$data, $[\text{'}$Country$\text{'}])$

covid$\_$world$\_$data

$\text{'}\text{'}\text{'}$

Now we can see our country data along with Covid$-19$ data for each country!

$\text{'}\text{'}\text{'}$ covid$\_$world$\_$data.describe$()\text{'}\text{'}\text{'}$

Note that we will likely have rows from our world data that did not match up with the WHO data. To see these, we can just

look for empty values in WHO columns $($since a left join would put empty values there$).$

$\text{'}\text{'}\text{'}$ covid$\_$world$\_$data$[$covid$\_$world$\_$data$[\text{'}$Total$\_$Vaccinations'$].$isnull$()]\text{'}\text{'}\text{'}$

Looking at these results, we can understand why the WHO may not have data on some of them. Some are not technically

countries $($like "Antarctica"$),$ some have ambiguous national status $($like "Taiwan"$),$ and some are just using different but

valid names $($like "United States" vs "United States of America"$).$

As a data scientist, think about ways that you could more efficiently join together these datasets.