1 00:00:00,600 --> 00:00:05,310 Now, let's start with links, simple linear regression model in Python. 2 00:00:08,200 --> 00:00:12,520 For that, first, we have to import Saaz Mother Lode API library. 3 00:00:12,730 --> 00:00:13,480 So we will write. 4 00:00:18,620 --> 00:00:20,410 Stucks mother thought EPA. 5 00:00:28,120 --> 00:00:28,510 s.M. 6 00:00:33,060 --> 00:00:35,520 If you remember what equation from. 7 00:00:37,140 --> 00:00:39,490 Linear regression was bikeways two. 8 00:00:39,930 --> 00:00:46,800 We don't know what must be done X by default, sex moderate do not include this be done time, which 9 00:00:46,800 --> 00:00:48,280 is also known as constant. 10 00:00:49,570 --> 00:00:57,910 Who at this consent to our data and cleared the X or independent variables for our model, will write 11 00:00:58,090 --> 00:01:02,940 X to at some doubt at consent. 12 00:01:10,050 --> 00:01:11,790 And then record will rate. 13 00:01:14,320 --> 00:01:14,880 B.F.. 14 00:01:16,860 --> 00:01:18,360 And they squared Rick, right? 15 00:01:18,460 --> 00:01:20,830 There were equitable, which is from them. 16 00:01:24,040 --> 00:01:32,620 Again, we are doing this to create our variable X and also to Erdan constant, which will be what we 17 00:01:32,620 --> 00:01:33,030 done or. 18 00:01:37,880 --> 00:01:43,410 The model will to create a modern object, which is L.M., right? 19 00:01:44,200 --> 00:01:48,720 Asymptote oil, this oil is a sense for ordinary Lisa the squid. 20 00:01:50,800 --> 00:01:57,250 And then record, we'll mention first of why a variable that is dependent variable, which is our price 21 00:01:57,270 --> 00:02:00,310 will, right, B.F. Price. 22 00:02:01,900 --> 00:02:06,020 And our ex, we have already defined politics to separate Elex. 23 00:02:07,120 --> 00:02:10,690 And after that, we will bring the mortals who will not big. 24 00:02:16,370 --> 00:02:19,820 But on this, I would object more to the object visit. 25 00:02:20,520 --> 00:02:22,620 And we have a model. 26 00:02:24,250 --> 00:02:27,390 To get the somebody of this mardon will, right? 27 00:02:28,790 --> 00:02:30,590 The modern name, which is L.M.. 28 00:02:33,360 --> 00:02:34,210 Not somebody. 29 00:02:38,300 --> 00:02:39,030 But on this. 30 00:02:42,490 --> 00:02:44,370 We will get somebody, somebody'll photo Martin. 31 00:02:46,170 --> 00:02:51,780 If you remember, our vivos equal to be done or less be done X. 32 00:02:56,530 --> 00:02:57,580 This is a lot to be done. 33 00:02:59,320 --> 00:03:00,730 And this is a little bit be done. 34 00:03:02,340 --> 00:03:07,850 You can also find the standard edit corresponding to each of these be done and be done. 35 00:03:09,790 --> 00:03:12,940 And we also have that P-value for these two barometer's. 36 00:03:14,620 --> 00:03:21,680 We also mentioned in the two Reflektor that be well, you should be less than one percent or five percent 37 00:03:21,980 --> 00:03:25,450 who have a significant relationship between X and Y variable. 38 00:03:26,110 --> 00:03:30,160 So here you can see that our B value is nearly equal to zero. 39 00:03:30,940 --> 00:03:36,280 You can see that that is a significant relationship between rule number and the price variable. 40 00:03:42,390 --> 00:03:47,880 On the top right side, you can also find ballou's of artist squid and just said artist squid. 41 00:03:49,140 --> 00:03:52,920 We mentioned in the two reelected that it should be clear to them point by. 42 00:03:55,180 --> 00:04:00,790 And you can see our artists go to value is near 2.5 and. 43 00:04:01,810 --> 00:04:07,780 This is pretty good, considering that we are only running good simple liquidation more than using one 44 00:04:07,780 --> 00:04:08,250 variable. 45 00:04:11,510 --> 00:04:15,650 Now there is another I hoodoo linear regression in Python. 46 00:04:17,080 --> 00:04:19,540 And this is the most commonly used. 47 00:04:20,630 --> 00:04:21,750 Linear regression matter. 48 00:04:22,100 --> 00:04:22,730 And Biton. 49 00:04:25,350 --> 00:04:28,120 In this matter, we will use a Skillern liberty. 50 00:04:30,810 --> 00:04:35,690 Well, right from a scale on the linear underscore model. 51 00:04:44,590 --> 00:04:46,300 Import linear regression. 52 00:04:56,150 --> 00:05:03,280 One thing you can notice here is that instead of using imported statement, which was import umpires 53 00:05:03,410 --> 00:05:11,360 and B and B or import ban does s speedy, we are using from Skillern Daughton in the unmodern import 54 00:05:11,360 --> 00:05:12,230 linear regression. 55 00:05:12,860 --> 00:05:18,920 The difference between these two is that if we are using imported statement such as for Mambi, you 56 00:05:18,920 --> 00:05:21,860 have to use Alias to use any of its function. 57 00:05:22,690 --> 00:05:28,100 But if you are using this first method, which is from a scale undoubtedly in the unmodern. 58 00:05:29,120 --> 00:05:34,160 You need not need to use Elías, you can medically use their functions. 59 00:05:37,490 --> 00:05:40,250 Just to give you some more information about Escalon. 60 00:05:40,730 --> 00:05:47,750 It's a free machine learning library for the Python programming languages, and it supports various 61 00:05:47,750 --> 00:05:50,960 classification regression and clustering algorithm. 62 00:05:52,390 --> 00:05:56,470 So going forward, what are your other advanced machine learning techniques? 63 00:05:56,500 --> 00:05:58,390 You will be using Escalon only. 64 00:06:01,530 --> 00:06:08,670 For Escalon, we first need to define our X and Y variable, so we will write Y equal to. 65 00:06:11,190 --> 00:06:14,600 Our very dependent variable is the price variable from the diaphragm. 66 00:06:15,330 --> 00:06:18,990 Will right be if it's got a record wheelwright price? 67 00:06:21,590 --> 00:06:27,350 Four x, one thing to note is what X variable will be a two dimensional array? 68 00:06:28,400 --> 00:06:33,720 So why variable was one dimensional umpiring I an exclude evil should be good. 69 00:06:33,740 --> 00:06:36,050 I mean, stun them by arrays will write X. 70 00:06:37,610 --> 00:06:38,050 Well, right. 71 00:06:38,850 --> 00:06:46,160 Well, to be if now to make it two dimensional, Larry, sort of just putting single this good record 72 00:06:46,280 --> 00:06:46,820 I will put. 73 00:06:47,040 --> 00:06:47,930 Whose good record? 74 00:06:49,050 --> 00:06:50,520 And inside it. 75 00:06:50,570 --> 00:06:52,480 I will write room number. 76 00:06:57,630 --> 00:07:01,510 Now, before forfeiting no more than we can first create a word. 77 00:07:02,580 --> 00:07:03,850 Linear regression objects. 78 00:07:04,320 --> 00:07:04,640 Right. 79 00:07:05,590 --> 00:07:07,230 Alem to equate to. 80 00:07:12,340 --> 00:07:13,130 The integration. 81 00:07:23,590 --> 00:07:25,450 We can execute this now. 82 00:07:25,920 --> 00:07:31,670 Allow them to the their regulation object early so we can for total murder, but I think elemental. 83 00:07:34,190 --> 00:07:36,890 But they say this. 84 00:07:37,190 --> 00:07:38,550 We can mention our X and Y. 85 00:07:38,780 --> 00:07:38,970 But. 86 00:07:43,600 --> 00:07:44,290 But on this. 87 00:07:47,910 --> 00:07:49,800 We have out our linear regression, Morten. 88 00:07:51,460 --> 00:07:54,520 Now to see the intercept and coefficient. 89 00:07:55,510 --> 00:08:03,200 Will right rain and record will rate L.M. to. 90 00:08:05,970 --> 00:08:07,300 Intercept int.. 91 00:08:11,330 --> 00:08:11,710 And put. 92 00:08:13,230 --> 00:08:17,080 And the school I know this will right them do, but. 93 00:08:23,100 --> 00:08:29,910 You can see we are putting this under school after intercept, and with these two are the attributes 94 00:08:30,120 --> 00:08:32,580 of what linear regression object? 95 00:08:34,550 --> 00:08:37,340 The intercept is minus they put one six five. 96 00:08:37,370 --> 00:08:38,260 This is what we've done. 97 00:08:39,580 --> 00:08:44,860 And I would be Dulan, which is the coefficient of X is nine point zero nine. 98 00:08:52,220 --> 00:08:57,860 Another thing I want to tell you is that if you want to need some more information about any of this 99 00:08:57,860 --> 00:08:59,150 function, you can write tell. 100 00:09:01,060 --> 00:09:08,290 You can get that information using what you can tell and then record, you can mention the object or 101 00:09:08,290 --> 00:09:09,940 the function you want the hell for. 102 00:09:10,120 --> 00:09:12,160 For example, if I want them to. 103 00:09:15,280 --> 00:09:16,240 I can run this. 104 00:09:18,170 --> 00:09:22,740 And I will get this help and documentation of this object. 105 00:09:28,570 --> 00:09:32,710 So here you can see this are the two attributes that we use. 106 00:09:33,460 --> 00:09:35,080 That is quiff and and that's up. 107 00:09:35,200 --> 00:09:37,450 You can read about it in more detail. 108 00:09:37,510 --> 00:09:37,930 See a. 109 00:09:41,790 --> 00:09:47,880 Similarly, if we want to credit the values of why, depending on the linear model we generated, will 110 00:09:47,880 --> 00:09:50,460 write Alem to predict. 111 00:09:52,560 --> 00:09:54,010 And then our equity will. 112 00:09:59,290 --> 00:10:00,540 We can also see this data. 113 00:10:01,090 --> 00:10:01,960 Another variable. 114 00:10:04,980 --> 00:10:09,580 If he does execute the tribunal, we will get this guy discredited values of why. 115 00:10:11,980 --> 00:10:14,760 Not to gloat over a regression line on Abdul. 116 00:10:15,810 --> 00:10:18,180 We'll use Seeman Library. 117 00:10:21,420 --> 00:10:26,730 If you remember, we plotted scatterplot using joint plots of seabourne liberty. 118 00:10:27,560 --> 00:10:29,010 So legs first. 119 00:10:30,140 --> 00:10:31,610 God has rindo. 120 00:10:33,000 --> 00:10:34,550 Of some, this joint block. 121 00:10:44,110 --> 00:10:49,960 Using help, you can get all the information you need to understand each and every function that you 122 00:10:49,960 --> 00:10:50,530 are using. 123 00:10:51,040 --> 00:10:54,690 So, for example, here you can see there are multiple parameters. 124 00:10:55,030 --> 00:10:58,040 We only use X, Y and data. 125 00:10:59,250 --> 00:10:59,860 But there are. 126 00:11:01,070 --> 00:11:01,520 Many more. 127 00:11:01,550 --> 00:11:02,340 But I mean, that's. 128 00:11:03,430 --> 00:11:08,050 Now, the Ford parameter, Urist, is off kind by default. 129 00:11:08,170 --> 00:11:11,050 This kind is set to scatterplot. 130 00:11:13,190 --> 00:11:17,410 To plot regression line, we need to pass this parameter as Greg. 131 00:11:17,890 --> 00:11:19,290 Greg, this central regulation. 132 00:11:20,240 --> 00:11:22,320 You can see there are other parameters as well. 133 00:11:22,580 --> 00:11:25,280 For height ratio or color of the. 134 00:11:27,000 --> 00:11:28,280 We are not discussing it, right? 135 00:11:28,320 --> 00:11:30,960 No, you can ride this on your own. 136 00:11:34,830 --> 00:11:40,500 If we want to plot the regression line, we'll write as soon as it's on. 137 00:11:40,770 --> 00:11:41,700 Go join Block. 138 00:11:48,050 --> 00:11:51,050 Then I would x variable X equal to. 139 00:11:52,640 --> 00:11:53,290 Be it. 140 00:11:58,090 --> 00:12:03,940 And then the squared record with will mention of a excludable, which is the. 141 00:12:10,170 --> 00:12:15,820 I will mention our way of variable, which is or more for the WNBA. 142 00:12:28,300 --> 00:12:31,430 Then kind equate to that, that. 143 00:12:41,170 --> 00:12:42,160 So if we run this. 144 00:12:49,070 --> 00:12:53,870 You can see we have the Lord be able to get a lot of praise and a number. 145 00:12:54,770 --> 00:12:57,470 And we can also see a line passing through this. 146 00:12:57,680 --> 00:13:00,130 This is a regression line that we generated. 147 00:13:03,770 --> 00:13:06,920 We have already discussed the slope and intercept of this line. 148 00:13:07,790 --> 00:13:09,380 But just to use your lies. 149 00:13:10,590 --> 00:13:13,490 This we can afford to be using seabourne liquidity. 150 00:13:29,750 --> 00:13:37,280 Another thing you can more this is that what we done was around nine and the meaning of 11 was by increasing 151 00:13:37,390 --> 00:13:39,350 the value of X. 152 00:13:39,920 --> 00:13:41,840 Much by will increase. 153 00:13:42,260 --> 00:13:45,170 So you can see by increasing one value of X. 154 00:13:45,890 --> 00:13:49,340 So suppose from four to five hour X. 155 00:13:50,610 --> 00:13:53,300 What vibe will increase from zero to 10? 156 00:13:53,780 --> 00:13:57,290 You can say that roughly it will increasing. 157 00:13:57,680 --> 00:13:59,480 It will increase by nine units. 158 00:14:03,330 --> 00:14:06,870 That's all you interpret the growth focus on this.