Here, we will use Python libraries GeoPandas and contextily to read and analyze datasets related to Airbnb listings in Barcelona. We will concentrate on visualizing data on$/$together with some maps. The datasets can be downloaded from this link $($copy$-$paste the link if it doesn't go there right away$).$ The size of the file is only around $4\mathrm{.}5$ Mb $.$ Unzip them into your working folder. In other words, your data should be sitting in the Data $\backslash$subfolder of your working folder.

There are two types of datasets. One is a $.$geojson file that contains the "geometry" of Barcelona neighborhoods $(73$ of them$).$ Others are listings of apartments$/$rooms available for rent on Airbnb at various times. These were downloaded from Airbnb archives and contain information about locations $($including which neighborhood$),$ prices, room types, and more.

As always with Python libraries, you will have to work with documentation, especially for visualization tasks. I went over some details in a class lecture, but you still might need to discover certain things by yourself.

This project requires creating a new environment and installing new packages as well as potentially reinstalling some of the old ones. This may require multiple attempts, perhaps updating Anaconda, and$/$or other packages. Start early!

Task $6.$

Read in 'listings$\_24.$csv$\text{'},$ 'listings$\_30.$csv$\text{'},$ 'listings$\_33.$csv$\text{'},$ and 'listings$\_45.$csv$\text{'}$ as well. The four csv files are Barcelona listings made at different times. For each of those find

The total number of listings for each neighborhood.

The "density" of listings $($total number divided by the area$)$ for each neighborhood.

The median price of listings for each neighborhood.

Note: You may need to "clean" the data so you don't end up with NaNs in your values $($e$.$g$.$ price$).$ If you find situations that produce NaN, deal with it the way you see fit and give a short explanation in a markup cell.

Task $7.$

Find the same information as in $6$ but this time only use listings for 'Entire home$/$apt$\text{'}$ room type.

The total number of listings for each neighborhood.

The "density" of listings $($total number divided by the area$)$ for each neighborhood.

The median price of listings for each neighborhood.

Task $8.$

Merge the results in $7$ about the "density" and "price" with the dataframe in task $3.$ In other words, you should have a dataframe with neighborhood information, plus $8$ more columns: density$\_24,$ density$\_30,$ density$\_33,$ density$\_45,$ price$\_24,$ price$\_30,$ price$\_33,$ price$\_45.($Suggestion: use pd$.$merge $($one$\_$dataframe, another$\_$dataframe, on$=$ 'sOME$\_$COLUMN$\_$NAME'$)$; or something similar, for example, with 'left$\_$on$=\text{'}$ and 'right$\_$on$=\text{'}$ instead of just $\text{'}$on$=\text{'}....$join will also work. See some examples here.$)$

Task $9.($I suggest do this problem LAST, it's not directly related to the ones above$)$

Create a horizontal bar plot of the Top $10($by "density"$)$ neighborhoods in such a way that bars show counts of each Room Type for each neighborhood $-$ stacked barplot. Use only one dataset, say listings$\_24.$csv$.$ First, figure out the top $10$ neighborhoods by "density". Then, for those $10$ neighborhoods only, find the counts of each room type. Then create a stacked bar plot for those $10$ neighborhoods, showing the counts for each room type. Something like this: