How do I remove columns from a pandas DataFrame?

Key Takeaways

hello and welcome back to my video Q&A series about the Panda's library in Python so the question for today is how do I remove columns from a panda's data frame great question uh so the panda data frame is of course the primary unit of uh data storage in the pandas Library it's like uh rows and columns so how do I remove a column uh as always we are going to pull up an example data set and uh but first we need to import pandas as PD and then uh we're going to pull up the UFO data set a data set of uh UFO sightings or reported sightings and so we'll use the read CSV function and we'll pass it the URL of the data set bit. Le slash UFO reports okay so this creates a pandas data frame and uh we'll just look at the head as a reminder of what is in this data frame they're just UFO reports it has uh five columns and in case you're curious you can use the shape attribute and see that there are 18,2 41 rows that's what this is telling us anyway uh let's pretend that we don't care about the colors reported column so we want to get rid of it okay how do we do that well let's uh we'll just use the drop method which is a data frame method so you say ufo. drop and then we're just going to pass it as a string the name of the column we want to drop so colors reported now there are two more things I'm going to add here the first is the axis so I'm going to say axis equals 1 and we'll talk more about axes in in another video but um in brief an axis is just kind of a direction okay so axis zero is the row axis um it kind of is a d like directionally like down is Axis zero I think of it is the row axis a AIS one is the column axis it's the one that kind of moves across and that movement will make more sense uh when we talk um more about it in a future video but axis equals 1 is the way I say to pandas I want to drop a column and its name is colors reported okay one more thing we're going to say in place equals true what does that mean uh it just means I want this operation to occur in place okay so uh we will run this and uh we will just rerun the head method and you will see that the colors reported column AKA series is now missing it's now gone okay so um you can actually drop multiple columns at once uh so I'll show you how to do that now so let's say I wanted to get rid of city and state I didn't care about those let's do ufo. drop and what you do is very similar except you just instead of passing a string you pass a list of strings okay so I'll say City uh whoops comma State okay so the first argument is that list the second argument is X is equal 1 third argument in place equals true we run that and UFO do head is now down to those two columns okay they are gone they are no longer part of the data frame and that's really it okay as always I like to end these videos with a bonus tip and the bonus tip is uh what if you want to remove uh rows instead of columns Well turns out you just use the drop method except you use axis zero cuz as I said axis one is the column axis axis zero is the row axis okay so uh so we're going to use ufo. drop and we know there's going to be something here and then we're going to say axis equals Zer and in place equals true but what is this first thing let's say I want to drop uh rows the first two rows because I don't care about UFO reports before 19 18 31 so I don't care about these two rows I don't want them in the data set so how do I refer to them um by name I actually the names are right here the names are zero and one okay just like these are the names of the columns these are the names of the rows they're uh officially called the index uh also called the labels and so I'm just going to pass a list and I'll just say Z comma 1 run that ufo. head and you you can see those two rows the 1930s have been dropped and if you really want to be sure that they're gone you can see that the shape uh the number of rows has decreased by two rows okay so uh one final note about this axis equals z is actually the default so technically I did not have to specify axis equals Z here uh however I always like to specify the axis just to be really clear about what I am doing okay so that's all for today thanks as always for joining me uh please click subscribe if you'd like to see more videos like this as always leave a comment uh telling me your question about this video or your Panda's question in general I will uh try to answer it maybe I'll even make a video okay so that's it uh thanks again and I will see you again soon