Pandas DataFrames

Quick reference

• 00:05 So in this video, I want to start to talk about data frames.
• 00:08 And data frames is where this starts to get fun, right?
• 00:10 Up until now we had just been dealing with sort of strange numbers and
• 00:14 they've been strung together in not a very visually appealing way.
• 00:17 But now we're going to be working with data frames.
• 00:19 And data frames, like I mentioned earlier, are like spreadsheets, they look
• 00:23 sort of like Excel, they're much more visually pleasing to deal with, right?
• 00:27 So let's just jump in and start building one.
• 00:30 So before we get started, I want to add one more thing up here.
• 00:33 Let's go from numpy.random
• 00:38 import randn, rand d.
• 00:42 Shift+Enter to run this, and this will allow us to create some random numbers.
• 00:46 So we're going to create a data frame, we need a bunch of random numbers for
• 00:49 that, and so that's all that is.
• 00:51 So let's create some data.
• 00:54 So let's go, my_data, and let's set this equal to randn,
• 00:59 and then (4,3).
• 01:03 And if we run this, we call my_data here, well,
• 01:06 you can just see we've created an array.
• 01:09 And it has one, two, three, four, rows, and one, two, three, columns.
• 01:15 So 4 and 3, right?
• 01:18 So that's cool, we just created some dummy arrays,
• 01:20 now your numbers will be different because it's randomly generating them, right?
• 01:24 So, okay, now we want, think of a spreadsheet,
• 01:27 it has columns and it has rows.
• 01:29 And usually those columns have headers and those rows have headers.
• 01:32 So let's create some of those.
• 01:34 Let's go, my_rows, and let's just create a Python dictionary.
• 01:39 And let's set these equal
• 01:44 to A and B and C and D.
• 01:48 Remember, we need four of them because we designated four there.
• 01:51 And let's create my_cols, my columns.
• 01:55 And again, this is going to be a Python dictionary.
• 01:57 So let's just call this one Monday, and we can go Tuesday, Wednesday.
• 02:04 And again, we're going to have three columns,
• 02:05 because we designated that right there, right?
• 02:08 So if we Shift+Enter, okay,
• 02:09 we've got these things, now we want to create an actual data frame.
• 02:13 And to do that, let's just name this my_df,
• 02:18 I'm using this my underscore thing, you can call these anything you want.
• 02:21 I just kind of like to call them, my whatever.
• 02:24 And set this equal to, now this is going to be a panda.DataFrame.
• 02:29 Remember, we're importing pandas as pd, so from now on,
• 02:32 we can do panda stuff, like referencing pd.
• 02:37 This is a function, and if we hit the Shift and the Tab key at the same
• 02:42 time on our keyboard, we can see what arguments get passed into here.
• 02:46 Now this is not a pandas thing, this is a Jupyter notebooks thing.
• 02:48 It'll always do this when you hit Shift+Tab on your keyboard.
• 02:52 You can see it wants us to pass the data, the index, which are the rows, and
• 02:57 the columns, and some other things we don't really care about.
• 02:59 So let's go ahead and do that.
• 03:01 So the data, we want this to be, my_data, just paste this in.
• 03:06 The rows, we want this to be my_rows, and the columns, we want this to be my_cols.
• 03:11 So if we Shift+Enter, nothing actually happens.
• 03:13 To run this thing, we have to call my_df, like this.
• 03:17 And when we do, boom, we get this nice spreadsheet looking thing.
• 03:21 And if we hover over things, right, it highlights.
• 03:26 Now, this is much,
• 03:26 much more sort of visually appealing than the NumPy arrays we looked at earlier.
• 03:31 And even the series we looked at in the last video,
• 03:34 we're starting to get somewhere here, right?
• 03:35 This is kind of cool.
• 03:36 Now, your display may look a little bit different.
• 03:38 I'm using the Firefox browser and this is how this looks in this browser.
• 03:42 Sometimes they have boxes around them,
• 03:45 just depending on your screen resolution or whatever, so very cool.
• 03:48 So, notice we passed in all of our stuff like this, right?
• 03:53 This just makes it easier, but we could actually pass in the data itself.
• 03:57 Let's actually change this from B to Z, just so
• 04:00 we can see the change when we run this.
• 04:03 So if we just copy this whole thing, instead of passing in the variable name,
• 04:06 we could just pass in the data itself.
• 04:08 Now, this is obviously not something you're usually going to want to do because
• 04:12 you'll often have a lot of data, right?
• 04:14 So it's hard to read if you try and pass in the data like this, but
• 04:17 we've just got a few, four things here, four rows, so we can do it like this.
• 04:22 And if we run this again,
• 04:23 you see now this has been changed to Z, because we made this change right there.
• 04:27 And very, very cool.
• 04:28 I'm just going to go ahead and change this back real quick.
• 04:32 And let's change this back to B, Shift+Enter to run this,
• 04:36 Shift+Enter to run this, Shift+Enter to run this, and we're back to B.
• 04:41 So those are data frames, right?
• 04:43 Not too complicated, pretty simple to make these.
• 04:46 Very cool to look at, you can do some things with them.
• 04:48 It's really fun.
• 04:49 So in the next video, we'll start to look at these columns.
• 04:52 We'll look at some things we can do with those, and that'll be in the next video.

Lesson notes are only available for subscribers.