Final task

[MieH-Dk]

For the final task, are we expected to make the dataframe for each year into a smaller dataframe only consisting of the observations for maximal temperature? If so - should we sort by higest temperature? And should the dataframe consist of only data from the 'ITE00100550' station or from all of the stations?

[joachimkrasmussen]

Hi Mie,

I'm assuming that you are asking about exercise 0.4.6. Here, you should simply use 'df_select' from before.

Best,
Joachim

[MieH-Dk]

Hi! No, I'm talking about Ex. 0.5.1.

[joachimkrasmussen]

Aha! Here you should use data from all stations, and you should not restrict your data to observations for maximal temperature.

/Joachim

[johankll]

Hi Joachim,

The assignment-hint says: "Don't forget to process the loaded DataFrame by using .iloc[] and changing the column names for each year as we previously did for the year 1863."

What are we exactly meant to do (i.e. what does it mean to "process the loaded DataFrame by using .iloc[])? I understand that we should modify the column-names as in Ex. 0.4.2... But should we also do any other modifications?

Also: is it necessary to use .iloc[] when changing column-names? I simply use .columns=[] which seems to work as well - am I missing something?

/Johan

[joachimkrasmussen]

Hi Johan,

What you are meant to do is the following:

1. Load data for each year
2. Keep only the first four columns for each year
3. Rename the columns appropriately
4. Store each data frame as an element in a joint list
5. Create a new joint data frame from this list
6. Reset the index

Does this make sense?

Regarding .iloc[]: It is not necessary to use .iloc[] when changing column-names. In fact, I don't think we are ever asking you to do so? Actually, using .columns=[] sounds perfectly appropriate :)

Best,
Joachim

[johankll]

Thanks a lot, Joachim.

You do actually ask us to use .iloc[] in the exercise hint.

/Johan

[joachimkrasmussen]

That's for the selection of columns (preprocessing) - not the renaming of columns:)

Best,
Joachim