Q$9-$ split$\_$dataset $(1$ pt$)$

Earlier, we defined a function select$\_$sample$()$ which allowed us to randomly sample a subset from our larger dataset.

Now, rather than specifying how many samples we want to randomly choose from a dataframe, let's take our entire dataframe and split it into thirds. We'll consider each third a different replicate of our experiment.

Write a function called split$\_$dataset, which will return a list of dataframes. Each dataframe in the list will be a random subsets $($without replacement$)$ of the input dataset.

Input$($s$)$:

dataset : DataFrame

n$\_$split : int, default: $3$

Output$($s$)$:

result : list of DataFrame$($s$)$

Procedure$($s$)$:

Use sample$()($from pandas, same as was used in Q$7)$ with the input parameters frac $=1$ and replace $=$ False. Store this in shuffled.

Use np$.$array$\_$split with shuffled and n$\_$split as its input parameters. Store this in result.

return result, which stores a list of dataframes