1 00:00:00,780 --> 00:00:07,650 First thing, AIDS, we're going to generate some data and some samples in this example, I'm going 2 00:00:07,650 --> 00:00:15,900 to use Sina's function and I want to train my neural network to give me the same output as seiners function 3 00:00:16,650 --> 00:00:17,760 before assorting. 4 00:00:17,760 --> 00:00:20,010 Let's just create a folder here. 5 00:00:20,010 --> 00:00:23,070 I already created a folder if I can find it. 6 00:00:23,190 --> 00:00:32,280 Matlab finds it's better to just keep all the data in one file and not confusing later with other files 7 00:00:32,280 --> 00:00:32,880 and folders. 8 00:00:32,910 --> 00:00:34,440 So let's start here. 9 00:00:34,440 --> 00:00:46,420 X equals linear space in space is a very dramatic progression and it can accept three different fields. 10 00:00:46,440 --> 00:00:49,170 The first field would be deliver a line. 11 00:00:49,890 --> 00:00:51,320 I'm going to set it to zero. 12 00:00:51,570 --> 00:00:54,420 The next field would be the upper line. 13 00:00:54,460 --> 00:00:58,230 Let's just go to two pi in Matlab. 14 00:00:58,230 --> 00:01:01,470 You can simply type P as a pie. 15 00:01:02,010 --> 00:01:04,350 And then how many samples do we need? 16 00:01:04,540 --> 00:01:06,420 Let's start with tweeny samples. 17 00:01:07,770 --> 00:01:08,610 Click, enter. 18 00:01:09,300 --> 00:01:17,220 So here I can see the generated data that we have 20 samples from zero to two PI. 19 00:01:17,230 --> 00:01:21,990 So it just divided from zero to two pies to any parts. 20 00:01:22,860 --> 00:01:25,390 And here I can see it now. 21 00:01:25,410 --> 00:01:27,960 Next one would be over output y. 22 00:01:28,230 --> 00:01:34,380 I will define Y as a sign of X again. 23 00:01:34,380 --> 00:01:40,260 If I just enter I can see the output of Y, let me do it again. 24 00:01:40,500 --> 00:01:49,470 I'm going to type ACLC to clean the board and I can use page up and page down to go to the code which 25 00:01:49,470 --> 00:01:51,140 I already wrote. 26 00:01:51,150 --> 00:02:01,440 So here I'm gonna go and generate X with a semicolon at the end to make it clear and then again generating 27 00:02:01,440 --> 00:02:03,560 Y this time again with semicolon. 28 00:02:03,900 --> 00:02:05,400 So I won't see the digits. 29 00:02:06,000 --> 00:02:13,040 Now it's time to plot plot X and Y and let's see what would be the output. 30 00:02:14,280 --> 00:02:17,640 So we're looking for another window to pop up. 31 00:02:18,210 --> 00:02:20,010 It's going to take a while. 32 00:02:20,160 --> 00:02:21,900 OK, here it is. 33 00:02:23,580 --> 00:02:24,660 Let's check it out here. 34 00:02:24,660 --> 00:02:26,970 We have twenty samples. 35 00:02:27,600 --> 00:02:30,510 If you zoom in, you can see this is not a curve. 36 00:02:30,540 --> 00:02:35,400 It just feeds a line and just connect all the samples here. 37 00:02:35,430 --> 00:02:40,230 You can see all the samples are just connected to each other. 38 00:02:40,230 --> 00:02:41,670 And this is not a curve. 39 00:02:41,820 --> 00:02:45,840 I want a neural network to fit a curve here. 40 00:02:46,230 --> 00:02:58,540 But for now, if you want to point out to your data, you can type plot X and Y and then here use A 41 00:02:58,620 --> 00:03:02,070 minus O inside, OK? 42 00:03:02,070 --> 00:03:08,910 And so I just want I'm going to use minus and then letter O close apprentices and let's check it one 43 00:03:08,910 --> 00:03:09,390 more time. 44 00:03:09,570 --> 00:03:11,970 This time I can see my samples better. 45 00:03:11,970 --> 00:03:15,180 And if you count them, if you don't trust you can just count. 46 00:03:15,180 --> 00:03:19,980 There are twenty samples that I have now let's back here. 47 00:03:20,610 --> 00:03:28,740 I already have the data I just generated twenty samples out of my output is sign yourself x now let's 48 00:03:28,740 --> 00:03:35,430 train over neural nets for and see how kind of neural networks give us the same output. 49 00:03:35,430 --> 00:03:38,340 Assign us of X for this purpose. 50 00:03:38,590 --> 00:03:40,110 Please go to your APS. 51 00:03:41,850 --> 00:03:47,580 This field is giving us several tools to use when you are working with artificial neural network. 52 00:03:47,790 --> 00:03:55,230 Here we have classification linear as we have curve fitting, we have neural net clustering. 53 00:03:55,770 --> 00:03:59,160 But let's start with neural net fitting. 54 00:03:59,160 --> 00:04:02,670 You can also click on a star next time. 55 00:04:02,670 --> 00:04:07,860 I can see it here at the top of my favorites, so just click on it. 56 00:04:08,370 --> 00:04:11,790 There is another way to open this page here. 57 00:04:11,790 --> 00:04:19,140 I can see an F two, so let's just click and close it and down and F to type it here. 58 00:04:19,770 --> 00:04:21,630 You can already see this window.