Numpy Operations

Quick reference

• 00:05 Okay, in this video I want to talk about NumPy operations,
• 00:07 specifically mathematical operations.
• 00:10 So we've got an array,
• 00:11 we want to do math to the individual elements inside that array.
• 00:15 How do we do that?
• 00:15 Well, there's basically a couple of ways we could do it.
• 00:17 We could use the array itself to do some math.
• 00:21 And we can use something called scalars.
• 00:22 Scalars are just basically numbers and we'll look at those in just a second.
• 00:26 So let's create an array, and I'm going to just call it my array.
• 00:30 Now in the last couple of videos, we just created some dummy np arrays, but
• 00:35 we can actually create them dynamically just by calling the arange function and
• 00:40 then passing in some ranges.
• 00:42 So if we want an array with from 0 to 10, we would just type in a range 0 11.
• 00:49 Now if we run this, nothing happens, but then if we come down here and
• 00:53 just do it like this, we can see here's our array.
• 00:56 So we just dynamically created some dummy data that we can play with.
• 01:00 So the first set of operations we're going to look at,
• 01:02 we call them array with array operations.
• 01:04 And we can just basically do any kind of math we want.
• 01:07 So we can go my_array plus, for instance, my_array.
• 01:11 And if we run this we'll see every element inside of our array gets added to itself,
• 01:16 because we're adding each array to itself.
• 01:19 So 0 plus 0 is 0, 1 plus 1 is 2, 2 plus 2 is 4,
• 01:23 3 plus 3 is 6, and all the way up to 10 plus 10 is 20.
• 01:28 We can do the same thing with subtraction, right?
• 01:31 So 0 minus 0 is 0, 1 minus 1 is 0 all the way up 0 all the way across, right?
• 01:39 We can do multiplication.
• 01:42 So 0 times 0 is 0, 2 times 2 is 4, 8 times 8 is 64,
• 01:47 all the way up to 10 times 10 is 100.
• 01:51 Now we can also do division.
• 01:52 Now this is kind of interesting.
• 01:53 Check this out.
• 01:54 When we do we get this warning.
• 01:56 And that's because what we're doing here is trying to take zero divided by zero.
• 02:00 We're getting this Nan.
• 02:01 That's a no warning.
• 02:02 So if you pull up a calculator and take 0 divided by 0 it's undefined right.
• 02:08 We remember that from math back when we're small children,
• 02:11 learning this stuff in school.
• 02:13 Also if you go any number say 5 divided by 0, you get, cannot divide by 0,
• 02:18 right, because that becomes infinity.
• 02:20 And normally if you're doing Python, and we can write down here and see.
• 02:24 We can take 0 divided by 0, and we'll just get an ugly error, right?
• 02:28 Division by zero not allowed, right?
• 02:30 Well, the nice thing about numpy arrays is, it allows us to do it.
• 02:35 It just gives us a warning, and then it returns a null object, right, NaN.
• 02:40 And we'll also see, if we tried to do 1 divided by 0 later.
• 02:45 With Python we get this huge error, with NumPy arrays,
• 02:47 we won't get an error at all, we'll just get a little warning like this.
• 02:51 And it will say infinity because any number divided by zero is infinity.
• 02:55 So that's interesting, you can also do exponents.
• 02:57 So double stars, and we get,
• 03:01 2 to the 2nd power of 5 to the 5th power, etc.
• 03:06 And you can see all of these big numbers put in there.
• 03:10 So those are arrays with array operations.
• 03:14 So we're just basically doing math using the array itself.
• 03:17 And the other way to do it is by using scalars like I mentioned at the beginning
• 03:20 of this video.
• 03:21 And scalars is just a fancy name for numbers, right?
• 03:24 So we can go my_array plus 10.
• 03:30 And that will just add 10 to every element in our array.
• 03:34 So zero plus 10 is 10, 1 plus 10 is 11,
• 03:38 5 plus 10 is 15, all the way up to 10 plus 10 is 20.
• 03:42 And again, same math operations we can do add, subtract, multiply, divide.
• 03:47 So we can subtract, we can multiply, right?
• 03:53 10 times 10 is 100.
• 03:55 We can divide.
• 03:58 Right now we get decimals because we're dividing.
• 04:01 Now, here, let's try this.
• 04:03 Remember I said when we divide by zero, that's infinity.
• 04:07 And a normal calculator won't let us do that.
• 04:09 But here we just get a warning and we get infinity, infinity, infinity,
• 04:13 infinity except the first one which is that NaN because that's again,
• 04:16 0 divided by 0, which is NAN number.
• 04:19 Very interesting and really, that's all there is to this NumPy operation.
• 04:24 So, this is data analysis, we're always going to want to do different types of
• 04:29 math on the elements inside of our NumPy arrays, and very,
• 04:33 very easy with NumPy using mathematical operations.
• 04:36 So that's all for this video.
• 04:37 In the next video, we'll look at universal functions.

Lesson notes are only available for subscribers.