WEBVTT 00:01.250 --> 00:06.330 In this video we are going to plot the data on our Titanic data frame with the method plot. 00:06.350 --> 00:11.730 But first of all let's import pandas and we import our Titanic data set. 00:11.750 --> 00:16.040 So here are the first five rows and some meta information. 00:17.990 --> 00:24.060 And if you want to create some plots and visualizations we first need to import the map plot lip library 00:24.350 --> 00:30.710 which is actually the most popular python library for creating plots and visualizations and actually 00:30.710 --> 00:35.140 as convention we import mad plot lip dubbed pi plot SBT. 00:35.870 --> 00:39.830 So a penalty is actually the most commonly used the previous Asian here. 00:39.860 --> 00:41.220 So let's do this. 00:41.330 --> 00:45.040 Now we want to plot to the numerical columns in our Titanic data frame. 00:45.050 --> 00:47.730 So we have some numerical columns here. 00:47.750 --> 00:55.280 We have this advice column the peak class column the H column space P parts and also the fact column 00:55.880 --> 01:03.050 and we can do this by simply calling here the plot method on our Titanic data frame so far Pena's data 01:03.050 --> 01:07.300 frames and pandas the areas that is the built and method the dot plot. 01:07.400 --> 01:11.360 And before we run the set here we have to also call the show method. 01:11.360 --> 01:11.900 So. 01:12.170 --> 01:15.800 So this is a direct map plot laptop type love method. 01:15.860 --> 01:20.070 And with that the show method we can actually display our graph our plot. 01:20.120 --> 01:25.460 So all of a sudden we create one or several plots of asylum and plug lip as the very last argument we 01:25.460 --> 01:28.120 have to call here the show method. 01:28.160 --> 01:34.690 So let's simply call here the plot method on our entire Titanic data frame and that's from the set here 01:36.930 --> 01:40.240 and here we can see that we actually see nothing. 01:40.680 --> 01:48.110 So by default the plot method creates here a line plot with on the x axis then the index. 01:48.120 --> 01:52.270 So here we have the range index from zero to eight hundred ninety one. 01:52.290 --> 01:58.800 So these are actually the passenger numbers and on the y axis we have actually these specific values 01:58.800 --> 02:03.690 and our numerical columns and for each column we have actually here a line plot. 02:03.780 --> 02:10.250 So we have six columns here and as we have here a six line plots more than one graph within one plot. 02:10.260 --> 02:16.500 That's quite hard to see anything and therefore let's go here inside the brackets. 02:16.580 --> 02:18.050 Let's shift tab. 02:18.060 --> 02:23.970 And here we can see we have a pair of meta subplots and by default it's set to forwards but we can set 02:23.970 --> 02:25.550 this also to true here 02:29.450 --> 02:33.200 and by doing so we're creating for each column a separate subplot. 02:33.260 --> 02:40.370 And still it's hard to see anything because the plot size is very small and that's an additional parameter 02:40.370 --> 02:41.960 here. 02:41.960 --> 02:48.580 And it's called the fixed size and we can determine the width and the height of the graph. 02:48.610 --> 02:50.220 So let's do this here. 02:50.240 --> 02:56.810 We have here fixed size and then we have to pass the then a top of the desired the width and the height 02:56.840 --> 02:57.650 of the graph. 02:57.650 --> 03:02.540 So for example you want to f a bit of twelve and a height of eight 03:05.700 --> 03:13.230 then we get to a larger plot and maybe you should further increase the size here to maybe 15 month 12 03:13.240 --> 03:16.800 of. 03:16.840 --> 03:19.030 So now it's much easier to see and analyze it. 03:19.540 --> 03:22.390 So for each column we have a separate subplot. 03:22.390 --> 03:28.570 So you have for the survival column the plus column the H column the best B column patch column and 03:28.570 --> 03:29.690 fact column. 03:29.840 --> 03:32.860 And by default all the subplots share actually. 03:32.890 --> 03:34.770 The x axis. 03:34.870 --> 03:41.230 So we have here at the very bottom one x axis the width of the Range Index from zero to eight hundred 03:41.230 --> 03:42.180 ninety one. 03:42.310 --> 03:46.110 But we can also create a separate x axis for each subplot. 03:46.120 --> 03:57.160 And that is the one parameter here and the parameter yes share X. So let's have a look here and by default 03:57.160 --> 03:58.910 it's set to true actually. 03:58.960 --> 04:06.020 So all subplots share one x axis but we can also set this to form. 04:06.050 --> 04:08.850 So let's do this. 04:09.940 --> 04:17.420 So we have these share experiments and we can set it to fours and we can see here at the bottom of each 04:17.420 --> 04:18.350 subplot. 04:18.380 --> 04:28.400 The x axis with the arranged index from 0 to 891 and we can see on the left hand side for each subplot 04:28.430 --> 04:30.500 actually the y axis. 04:30.530 --> 04:33.260 And here we have a specific y axis. 04:33.260 --> 04:34.310 For each subplot. 04:34.310 --> 04:40.290 So for example here if this survives column we have values between 0 and 1. 04:40.400 --> 04:47.210 So you can see here it's isolating between 0 and 1 then here for the peak class column we have the values 04:47.240 --> 04:55.670 1 2 and 3 then for the H column we have values between 0 and 80 and the line plot like this is actually 04:55.670 --> 04:58.130 good to see in the distribution of values. 04:58.250 --> 05:02.540 So the majority of ages are yeah between 20 and 40. 05:02.780 --> 05:08.750 But we have some outliers here around zero and some outliers around the 75 80. 05:09.650 --> 05:13.950 And specifically when we have time serious data with a daytime index. 05:14.030 --> 05:17.530 So we have been on the x axis a daytime index. 05:17.540 --> 05:25.040 So for example starting from the year 2010 onto two thousand and fifteen and specifically in this case 05:25.040 --> 05:29.750 a line plot is very intuitive to see the evolution of some factors over time. 05:29.900 --> 05:36.860 So for example stock prices or weather data but also you have a non time serious data a line plot to 05:36.860 --> 05:37.390 the max. 05:37.390 --> 05:44.060 That definitely sentence and here in this FSP column we can see that the majority of fair use are here 05:44.060 --> 05:45.010 actually 1. 05:45.620 --> 05:49.990 So most of the passengers have one sibling or spouse onboard. 05:50.360 --> 05:57.860 And also you have for the parents and children the most frequent values are between 0 and 2 and also 05:57.860 --> 06:05.000 you in the fact column we can see that the most frequent fast are actually quite low year between 0 06:05.000 --> 06:06.380 and 20. 06:06.530 --> 06:08.300 However there are some outliers here. 06:08.300 --> 06:16.040 So we have some outliers onto four or five hundred dollars and we have seen on the left hand side so 06:16.040 --> 06:17.120 fights subplot. 06:17.180 --> 06:22.700 We have a different scale for the y axis but we can also change this year. 06:22.760 --> 06:24.920 So yes that apparently does share y 06:27.990 --> 06:29.240 so we have Y 06:32.460 --> 06:35.160 and by default it's set to fall so far a subplot. 06:35.170 --> 06:41.480 We have a specific and appropriate y axis but we can also set this to a true 06:44.030 --> 06:44.640 a.. 06:44.650 --> 06:45.640 For a subplot. 06:45.640 --> 06:49.500 Actually these same y axis with the same scale. 06:50.460 --> 06:55.230 And as we have here the largest scale at the fact column. 06:55.290 --> 07:00.210 So the sad y axis is actually most appropriate here for the fact column. 07:00.210 --> 07:03.730 So here we have values between zero and 500. 07:03.750 --> 07:09.310 However for the patch column a scale between zero and five hundred does not really make sense. 07:09.310 --> 07:11.300 So here we actually see nothing. 07:11.460 --> 07:17.130 And therefore in this case that definitely makes sense to set these shall y Wyatt parameter to fours 07:20.810 --> 07:26.300 and actually if you are only interested in the H column and if you only want to visualize that the H 07:26.300 --> 07:33.440 column you can also apply the plot method only on the H column so let's do this let's copy the code 07:36.890 --> 07:42.770 and instead of using the plot method on the entire Titanic that afraid to use it only on the H column 07:44.060 --> 07:48.070 and if you only use it on the edge column we do not have to define me as subplots. 07:48.140 --> 07:55.580 So this is not necessary and maybe we make the figure a size a bit smaller here because we have only 07:55.580 --> 07:58.660 one subplot. 07:58.880 --> 08:07.020 And also we do not need to do apparently does share X and shall y so let's plotted the graph. 08:07.220 --> 08:14.180 And here we have actually a line plot only for the H column so we can't see the x axis that the index 08:14.180 --> 08:16.390 letters from 0 to 890 is. 08:16.670 --> 08:23.930 And we can see on the by axis values between zero and 80 and the majority of fair use are here between 08:24.050 --> 08:31.810 20 and 40 and we have some very young passengers that are only a bit older than the serial Yes loads 08:31.810 --> 08:38.210 so babies and we have some quite old peasants as above 70 and until 80. 08:38.530 --> 08:39.050 All right. 08:39.050 --> 08:45.770 This was the first video on data visualization and in the next video we learn how to customize our plot 08:45.770 --> 08:46.130 here. 08:46.130 --> 08:48.170 So I hope to see you there by.