1 00:00:00,630 --> 00:00:07,140 How that continue with our project and this video, I want to talk about some visualizations all. 2 00:00:07,920 --> 00:00:10,550 But first we have to explain what does it mean over there? 3 00:00:11,370 --> 00:00:12,300 So Sigmund's. 4 00:00:13,370 --> 00:00:19,400 At the Whiskas represent the lower and upper Whiskas, or as you say, we have lower and upper whisker, 5 00:00:19,910 --> 00:00:27,440 somebody forward a whisker where I stand up to one point five times in the entire Koltai range from 6 00:00:27,440 --> 00:00:31,070 the top or bottom of the box to the furthest. 7 00:00:32,280 --> 00:00:41,070 Think that is done so in this way, therefore, equally populated ranges delineated by courthouses are 8 00:00:41,580 --> 00:00:47,520 graphically represented, sort of like the plot in Python, we can use the library. 9 00:00:48,120 --> 00:00:59,490 So the library is a python to the library that produces publication quality features in a variety of 10 00:00:59,490 --> 00:01:03,870 hardcopy formats and interactive environments across all platforms. 11 00:01:04,300 --> 00:01:09,290 So the library try to make easy things and things possible. 12 00:01:09,750 --> 00:01:18,030 You can generate histogram power, Spectra apologized, Energis Scatterplot and so on, which are a 13 00:01:18,030 --> 00:01:28,270 few of the pilot functions consist of a collection of commands type function that might not work, such 14 00:01:28,290 --> 00:01:28,920 as Matlab. 15 00:01:28,920 --> 00:01:36,900 So each pilot function might change to a figures such as creating a figure, creating a plotting area 16 00:01:36,900 --> 00:01:42,940 in a figure, plotting some lies in a plotting area, decorating the plot with labels and so on. 17 00:01:43,470 --> 00:01:47,610 So we will do the primary in this project. 18 00:01:47,610 --> 00:01:50,580 So we will you in part. 19 00:01:52,540 --> 00:02:01,380 Not as party, so the available data in the bands that are kept from us, so for this reason we can 20 00:02:01,380 --> 00:02:10,530 use data from loss of function so they function megabuck block from the data frame columns, which are 21 00:02:10,530 --> 00:02:13,530 optionally drop by some other columns. 22 00:02:14,220 --> 00:02:16,260 So now let's create some. 23 00:02:24,130 --> 00:02:26,410 Equa later, Don. 24 00:02:27,970 --> 00:02:28,810 Box blog. 25 00:02:31,520 --> 00:02:32,120 Column. 26 00:02:34,260 --> 00:02:36,780 Equal pay names. 27 00:02:37,870 --> 00:02:40,480 And the data show. 28 00:02:47,740 --> 00:02:50,590 And now we have our box bloss. 29 00:02:59,750 --> 00:03:02,920 So it's very small to say so. 30 00:03:06,180 --> 00:03:09,240 We cannot know several variables have our eyes. 31 00:03:10,270 --> 00:03:14,280 With the Green Beret variable being the one that has the largest number. 32 00:03:16,020 --> 00:03:16,770 So. 33 00:03:18,300 --> 00:03:25,980 We will need to look on how to handle the problem, so first of all, from the. 34 00:03:26,890 --> 00:03:34,110 Analysis, we can say that I predict are many of us can create problems rather than just keeping their 35 00:03:34,720 --> 00:03:43,180 hands so we can identify which of the available predictors are most correlated with the response variable. 36 00:03:43,780 --> 00:03:50,770 So a standardized measurement of the relationship between two variable is that presented by the correlation, 37 00:03:51,190 --> 00:03:54,520 which can be calculated starting from Cauvery. 38 00:03:54,520 --> 00:04:02,250 And so in Python, correlation and causation are calculated by the bonds that the rendered call function. 39 00:04:02,770 --> 00:04:05,450 This can be a wise correlation of columns. 40 00:04:05,860 --> 00:04:07,950 I slowly and I and no values. 41 00:04:08,230 --> 00:04:13,810 So there are three available which are so they are dream of. 42 00:04:16,110 --> 00:04:17,370 Then right on take. 43 00:04:18,360 --> 00:04:20,280 They are three. 44 00:04:21,170 --> 00:04:22,220 Matthau's of. 45 00:04:23,300 --> 00:04:28,940 Call Rabelaisian, call aviation. 46 00:04:36,000 --> 00:04:37,470 So the first one is. 47 00:04:41,640 --> 00:04:41,930 John. 48 00:04:43,510 --> 00:04:44,460 What is the. 49 00:04:45,590 --> 00:04:52,100 Standout correlation, high efficiency. 50 00:04:54,560 --> 00:04:55,670 The next one, A. 51 00:04:57,140 --> 00:04:59,150 Kendall tão. 52 00:05:02,160 --> 00:05:03,300 Correlation. 53 00:05:06,040 --> 00:05:09,880 Call a and the last one is. 54 00:05:12,540 --> 00:05:13,560 Spearmen. 55 00:05:15,170 --> 00:05:15,920 With a. 56 00:05:20,810 --> 00:05:21,410 So. 57 00:05:24,060 --> 00:05:26,550 This one is Kendall. 58 00:05:30,350 --> 00:05:30,830 Tal. 59 00:05:34,850 --> 00:05:35,970 And on the. 60 00:05:37,990 --> 00:05:38,950 Spearmen. 61 00:05:40,430 --> 00:05:42,620 Right, correlation. 62 00:05:44,480 --> 00:05:51,260 So remember, the correlation coefficient apply to random variable is a measure of their linear dependent. 63 00:05:52,280 --> 00:05:53,900 So now let's calculate at. 64 00:05:56,370 --> 00:05:57,390 And, of course, Al. 65 00:05:58,760 --> 00:06:02,210 And I think this is one regarding what is. 66 00:06:05,970 --> 00:06:14,520 So call data, equal data scale, don't call. 67 00:06:16,680 --> 00:06:17,850 That made her. 68 00:06:19,770 --> 00:06:20,460 Iko. 69 00:06:22,140 --> 00:06:22,860 Passan. 70 00:06:27,080 --> 00:06:35,830 So to display all the data from column, we can also underscore quantize with one or more options for 71 00:06:36,500 --> 00:06:37,820 that rise and call for that. 72 00:06:39,320 --> 00:06:44,210 So what we did on option. 73 00:06:47,880 --> 00:06:48,720 Sudden contacts. 74 00:06:50,210 --> 00:06:50,990 And. 75 00:06:52,290 --> 00:06:53,160 They supply. 76 00:06:56,730 --> 00:06:57,080 Dorte. 77 00:06:58,650 --> 00:07:05,880 Not under his control, so he has to be careful not I don't want to make any mistake. 78 00:07:06,690 --> 00:07:12,180 So this time marks and then on and then display McCollom. 79 00:07:13,440 --> 00:07:16,180 Here, so I get it from here. 80 00:07:16,200 --> 00:07:22,920 So it must be easier for me and then come up with a. 81 00:07:24,830 --> 00:07:25,460 Call. 82 00:07:29,990 --> 00:07:31,140 Data, though. 83 00:07:33,770 --> 00:07:35,180 SCHIP is one. 84 00:07:37,750 --> 00:07:40,870 And then in here, I will bring. 85 00:07:43,190 --> 00:07:47,630 Call it a on the sale. 86 00:07:48,920 --> 00:07:50,700 And we got our results. 87 00:07:55,470 --> 00:08:04,530 So due to a large number of variables, the matrix is not easily interpretable, so overcome to overcome 88 00:08:04,530 --> 00:08:12,790 this inconvenience, we did a lot of correlation coral, coral or so a coral or issacharoff. 89 00:08:13,110 --> 00:08:15,390 So let's try it because I my. 90 00:08:17,860 --> 00:08:22,570 Pronoun is wrong, so we need to plot call. 91 00:08:23,900 --> 00:08:25,990 Let's call the. 92 00:08:27,510 --> 00:08:28,210 No prob. 93 00:08:29,870 --> 00:08:39,380 So Calderon is a rap about corre correlation matrix, so it's easier for it to highlight the most correlated 94 00:08:39,380 --> 00:08:42,180 variables in the data variable. 95 00:08:42,680 --> 00:08:48,270 So in this plot correlation coefficient r color according to the value. 96 00:08:49,720 --> 00:08:50,470 So let. 97 00:08:51,980 --> 00:08:52,480 Logit. 98 00:08:57,560 --> 00:09:01,040 So this is our. 99 00:09:04,240 --> 00:09:09,910 I'll call the nation matrix. 100 00:09:17,510 --> 00:09:18,140 So let. 101 00:09:19,390 --> 00:09:26,290 Right, I'm caught for that, so it's very easy to put it on. 102 00:09:27,770 --> 00:09:29,720 My show. 103 00:09:31,870 --> 00:09:38,380 Call data and then pretty darn X. 104 00:09:41,850 --> 00:09:42,510 Thanks. 105 00:09:43,920 --> 00:09:44,520 So. 106 00:09:46,870 --> 00:09:48,500 It should be Reisz. 107 00:09:50,560 --> 00:09:52,450 And then, Glenn. 108 00:09:54,980 --> 00:10:00,230 And then call it the don't. 109 00:10:01,800 --> 00:10:03,900 Call Plum's. 110 00:10:09,310 --> 00:10:10,840 And then. 111 00:10:14,960 --> 00:10:15,530 Call. 112 00:10:19,060 --> 00:10:19,470 The. 113 00:10:24,410 --> 00:10:25,010 Call them. 114 00:10:37,390 --> 00:10:46,220 So let's close our project and then we'll continue to be high stakes raised. 115 00:10:48,380 --> 00:10:50,870 And then the last one will be. 116 00:10:52,780 --> 00:10:57,210 Partido Kollar by. 117 00:10:59,110 --> 00:11:01,870 And they people, they don't show. 118 00:11:04,270 --> 00:11:07,450 So let me explain a bit of this quote. 119 00:11:09,380 --> 00:11:17,630 So, as you know, that is a real correlation matrix, so it's is very useful to highlight the most 120 00:11:17,630 --> 00:11:20,490 correlated variables in a data table. 121 00:11:21,170 --> 00:11:27,530 So in this plot correlation coefficient, Akela, according to their value correlation matrix, can 122 00:11:27,530 --> 00:11:32,290 be riada according to the degree of association between variables. 123 00:11:32,630 --> 00:11:39,030 So we can plot a over in using the Marlott not by myself function. 124 00:11:39,590 --> 00:11:42,290 So in here, as you say, we use the muscle function. 125 00:11:42,680 --> 00:11:47,210 This display a data frame as a matrix in the window. 126 00:11:47,600 --> 00:11:55,040 The region, he said, had upper left hand corner and rose from first dimension of your right eye display 127 00:11:55,040 --> 00:11:55,970 horizontally. 128 00:11:56,210 --> 00:12:00,320 The aspect ratio of the figure we know is that right? 129 00:12:00,330 --> 00:12:04,370 And this would make an extensive shot on our row figures. 130 00:12:04,940 --> 00:12:09,260 So thick label as far as applied on the top. 131 00:12:10,790 --> 00:12:13,880 So from the court, we're going to say that. 132 00:12:15,330 --> 00:12:25,320 We come to know the payout a lot and pay out a lot, White said the current location and labels of the 133 00:12:25,350 --> 00:12:28,150 ISIS and why ISIS pay out they don't. 134 00:12:28,230 --> 00:12:29,730 Colaba at. 135 00:12:30,920 --> 00:12:36,620 Colaba to applause And finally here, the show is sort of like. 136 00:12:38,060 --> 00:12:40,760 So now that rhizome. 137 00:12:43,410 --> 00:12:44,730 So this one is. 138 00:12:51,740 --> 00:12:53,600 So that cries out. 139 00:12:53,960 --> 00:12:57,890 So this one will set the current date. 140 00:12:58,840 --> 00:13:05,230 Locations and labels, they ask. 141 00:13:06,010 --> 00:13:10,480 Access and y axis. 142 00:13:12,400 --> 00:13:14,380 And why access? 143 00:13:15,600 --> 00:13:18,130 And they don't Colaba at all. 144 00:13:18,600 --> 00:13:24,240 So it is one Ettakatol and Radioshow and then let Grundler sell. 145 00:13:25,350 --> 00:13:29,280 And I got on error. 146 00:13:31,570 --> 00:13:35,710 What is it involve Shinda, so. 147 00:13:40,930 --> 00:13:53,070 So either we call added to a liberal court and then I misspelling it in here and also rewrite, of course, 148 00:13:53,440 --> 00:13:54,680 it had quite a few times. 149 00:13:54,910 --> 00:13:56,060 And this is our blog. 150 00:13:56,710 --> 00:14:03,610 So as I said, that in the existing relationship between the response variable and the predictors, 151 00:14:03,760 --> 00:14:07,800 we will only analyze the last line of the correlation matrix. 152 00:14:08,320 --> 00:14:18,700 So in it, we can say that predictors that are the most closely related, namely I'm ousted and Patriota, 153 00:14:19,150 --> 00:14:24,590 indeed this variable have colors that I brought the extreme of the color label. 154 00:14:25,150 --> 00:14:32,510 So this is the last that I am in this variable hair color approach, the extreme of the color label, 155 00:14:32,770 --> 00:14:36,330 the different color is due to the positive and negative correlation. 156 00:14:36,940 --> 00:14:41,380 And so in the color label, in the right hand part of the plot. 157 00:14:42,070 --> 00:14:44,080 And that is only this way. 158 00:14:44,680 --> 00:14:48,010 I hope you enjoy it and I will see you in the next one.