Numpy Universal Functions

Quick reference

• 00:04 Okay, in this video I want to look at some universal functions that are built into
• 00:09 NumPy that allow us to do different operations on our arrays.
• 00:13 So we got the same array from our last video, we called it my_arr.
• 00:17 And we just set this equal to this np.arange.
• 00:20 So we generated some dummy data for us from 0 to 10.
• 00:24 Well, 0 to 11, but not counting 11.
• 00:26 So here we have that.
• 00:28 Now, like I said, NumPy comes with some universal functions built in,
• 00:32 that we can play with and use.
• 00:34 So let's just start looking at these.
• 00:37 If we wanted the square root of each item, we could
• 00:42 go np.sqrt stands for square root, and then just pass in our array.
• 00:47 And then boom, we get the square root of each of these items.
• 00:51 So the square root of 9 is 3.
• 00:54 You see right there, 3.
• 00:56 So very cool.
• 00:58 If we want the absolute value of each one we call absolute.
• 01:04 And these are all positive numbers so we get just the same ones.
• 01:09 If we went negative 10 here, and then run this again we get some negative numbers.
• 01:17 Well, if we run the absolute value of that we would see the absolute value
• 01:21 of negative 10 is 10, etc.
• 01:23 So that's kind of cool.
• 01:25 Go ahead and change this back.
• 01:26 We can call the exponent on each one.
• 01:30 Right, remember in the last video we did that by going my_arr times times,
• 01:38 for instance, my_arr or whatever.
• 01:44 We can also use a function to do that.
• 01:45 So that's cool.
• 01:48 And there's just a few of these I want to talk about and
• 01:49 then I'll show you where you can find a list of these.
• 01:51 We can find the max, what is the highest number, the max number?
• 01:55 Well, that's 10, right?
• 01:57 Interesting.
• 01:58 We could go min and that becomes 0.
• 02:01 Often you want to find what the highest number and the lowest number is,
• 02:04 the max and the min, so that's an easy way to do that.
• 02:08 We can find the sign, are these positive or negative, right?
• 02:12 So, 0 is neither positive nor negative, the rest of these are all positive,
• 02:16 so we get a 1 for that, all right?
• 02:18 If we came up here and went -10 again, and then run this again.
• 02:25 And then called this, we get negative, negative, negative, negative, for
• 02:30 all these negative numbers, so that's interesting.
• 02:33 Let me change this back.
• 02:36 We can do trig.
• 02:37 Remember trigonometry?
• 02:39 So you can call the sine or the cosine, right?
• 02:45 We could call logarithm, and
• 02:47 then we get an infinity because the log of 0 is infinity.
• 02:51 But for the rest of them, we get the logarithm from each one.
• 02:53 So very, very cool and very easy to use just by passing in
• 02:58 your array into whatever function, universal function you want to use.
• 03:03 And I mentioned there's a bunch of these you can draw on.
• 03:05 So let's open another web browser, and this is the official documentation for
• 03:11 NumPy and it's just docs.scipy.org/doc/numpy/reference/ufuncs,
• 03:18 universal functions.
• 03:21 And this is the page for the universal functions.
• 03:23 Then if you scroll down towards the bottom, you can read all about this if you
• 03:27 want but the thing we really care about is we get down here and
• 03:30 we start seeing a list of all of these functions.
• 03:32 Now we did math in the last video, math operations.
• 03:35 You can also use each one of these universal functions to do your math too.
• 03:38 I figure, I feel like using the math we did in the last video is easier than
• 03:43 calling the add function if you want to add, but
• 03:46 you can always do that if you want.
• 03:48 So here are all of these.
• 03:52 So there's a bunch of math operations, right?
• 03:55 There's the square root we just did, you can find the square, reciprocal,
• 03:59 all kinds of cool things.
• 04:00 There's all the trig functions you would expect, sine, cosine, tangent, etc, etc.
• 04:06 Bit-twiddling, we probably won't look into any of this stuff.
• 04:10 Comparison functions and some other things, floating functions.
• 04:14 So very cool.
• 04:14 So check this resource out if you want to refer back to this in the future.
• 04:19 But pretty easy to use these universal functions with NumPy.
• 04:22 So that's pretty much all we're going to talk about with NumPy.
• 04:25 We're going to use NumPy throughout the rest of this course a lot.
• 04:29 And from time to time, I might show you some new thing with NumPy in
• 04:33 order to help us use it a certain way going forward.
• 04:35 But for the most part,
• 04:36 that's really all I want you to wrap your heads around at this moment.
• 04:39 Numpy arrays, they're just basically Python lists only they hold more data.
• 04:45 You can do math operations to them.
• 04:47 You can use functions on them.
• 04:49 And really, that's all we need to get started.
• 04:51 So in the next section, we're going to jump into Pandas.
• 04:53 I'll tell you all about it and show you how to install it and get it set up.
• 04:55 And that'll be in the next video.

Lesson notes are only available for subscribers.