I have a large dataset of certain events for the my research industry, organized in a dataframe as follows. Each event has an event type (str), a year of the event (int), event size (int) and an event location (str).
An example dataframe is structured below, with event types 'A', 'B', 'C', or 'D' and event locations 'CA', 'TX', 'NY'.
| Event Number | Event Type | Year | Size | Location |
|---|---|---|---|---|
| 1 | A | 2014 | 1000 | CA |
| 2 | B | 2014 | 1000 | TX |
| 3 | C | 2014 | 456 | CA |
| 4 | C | 2014 | 675 | NY |
| 5 | B | 2014 | 567 | TX |
| 6 | A | 2014 | 765 | CA |
| 7 | C | 2014 | 1000 | NY |
| 8 | B | 2014 | 675 | TX |
| 9 | D | 2015 | 3424 | NY |
| 10 | A | 2015 | 567 | TX |
| 11 | A | 2015 | 435 | CA |
| 12 | C | 2016 | 45 | CA |
Now, I want to plot a heatmap of event type vs year. i.e., a heatmap with year on the x axis, event type on the y-axis, and a heat color representing a count of how many of those types of events happened in that year. The resulting matrix for the above table would look something like this:
| Event Type | 2014 | 2015 | 2016 |
|---|---|---|---|
| A | 2 | 2 | 0 |
| B | 3 | 0 | 0 |
| C | 3 | 0 | 1 |
| D | 0 | 1 | 0 |
I have looked into using seaborn but I am not sure how to approach this sort of 2D histogram.
How would I go about it if I also wanted to plot a heatmap of location vs event type (2 strings)?
Thanks!
