WEBVTT 00:00.630 --> 00:06.080 In this video we learn how to sort data frames by column values and by index labels. 00:06.080 --> 00:14.070 So first of all we import pandas and we work with the Titanic's dataset and we will have a look at the 00:14.070 --> 00:16.620 first five frozen the last five rows 00:19.730 --> 00:23.160 and what we actually could do we could select a single column here. 00:23.180 --> 00:27.330 The H column and we can start the H column by the values. 00:27.340 --> 00:31.400 So let's first of all let's check here. 00:31.400 --> 00:39.370 Our H column so we get here a pen that's serious and we can see here that the values now append a serious 00:39.370 --> 00:41.000 set from the same order. 00:41.000 --> 00:49.090 Then in our original data frame starting from twenty to thirty eight twenty six and so on and with the 00:49.140 --> 00:55.660 salt values method we can actually start our pandas series. 00:55.700 --> 01:01.910 So now we can see here that our pandas serious starts with the peasant so that is 0 point for 2 years 01:01.910 --> 01:06.320 old and then we have four point six seven years. 01:06.320 --> 01:13.830 And on the left side we can see the indexed labour of the peasants in our original data frame and we 01:13.970 --> 01:21.120 the assault various method we not only can sort pandas serious but also pandas data frames. 01:21.380 --> 01:28.400 So the software use method does not only exist for serious but also for data frames so we can apply 01:28.400 --> 01:31.780 here the method directly to our data frame. 01:31.980 --> 01:38.300 And now if you want to start our complete data frame our complete rose by the H column we have to pass 01:39.750 --> 01:46.290 the label off the H column here in quotation marks to the method to salt values so let's check this 01:46.290 --> 01:46.760 out here 01:49.780 --> 01:55.870 and here we can see that we have the complete data frame sorted by the H column so starting with peasants 01:55.900 --> 02:02.620 a number eight hundred and three year that is 0 point for two years old and we can see here that by 02:02.620 --> 02:06.760 default we have from ascending order so starting from the youngest the peasants are. 02:07.150 --> 02:14.620 So let's check again our Titanic data set and print the first five rows and you can see here that our 02:14.800 --> 02:22.730 original data frame is not sorted so there must be obviously also a in place parameter in the sort that 02:22.730 --> 02:27.960 use method then let's check the parameters of this sort various method by pressing shift tab. 02:28.100 --> 02:32.000 And here we can see that we can start by the values along either axis. 02:32.030 --> 02:37.670 So we have to pass for the parameter by we have to pass that the column or the columns by which we want 02:37.670 --> 02:39.100 to sort our data frame. 02:39.500 --> 02:45.650 And this indeed is the in-place parameter which we have just set to true if you want to if you want 02:45.650 --> 02:52.950 to change our original data frame so let's set it to false here and we have the ascending parameter 02:53.010 --> 02:58.140 that we can define whether we want our data frame started in ascending or descending manner. 02:58.350 --> 03:05.940 So if you put here the parameter defaults we have a descending order and we have to define the axis 03:05.940 --> 03:07.540 along which we want to start. 03:07.590 --> 03:11.520 And in this example here we want to sort of a rose by the H column. 03:11.550 --> 03:12.780 So we want to sort our 0. 03:12.790 --> 03:17.050 So we have to pass out those 0 Art Index here and Secretary yet 03:21.040 --> 03:28.420 so here we have now our Titanic data frame sorted by the H column and then descending manner so starting 03:28.420 --> 03:38.310 from the oldest passenger and we can also change the Sierra again back to true so true as the default 03:38.310 --> 03:40.460 parameter here. 03:40.490 --> 03:42.760 So now we have an ascending order here. 03:42.980 --> 03:48.620 And once we are happy with our sorting we can then set the in-place parameter to true 03:52.490 --> 04:00.330 and let's check our data friends that we should have a new data frame here sorted by the H column and 04:00.360 --> 04:02.640 starting from the youngest passenger. 04:02.640 --> 04:05.530 So in a ascending order. 04:05.580 --> 04:12.060 All right so and b not only can thought our data frame by one column but also by many columns and therefore 04:12.090 --> 04:20.130 we can pass a list of column labels so we have passenger class sex and age and we set here the ascending 04:20.130 --> 04:21.550 parameter to fours. 04:21.560 --> 04:23.130 So let's see what we get here 04:27.270 --> 04:29.950 and what actually pendant does here at sorts. 04:30.030 --> 04:36.380 The data frame first by passenger class will starting from the first element then by sex. 04:36.420 --> 04:42.960 The second element and then by age and the third element in our list and we have a descending order 04:42.960 --> 04:45.210 so starting from height values. 04:45.420 --> 04:47.810 So let's have a look here at the passenger class column. 04:47.820 --> 04:50.390 So we are starting with class 3. 04:50.610 --> 04:56.550 So the highest value of the classes and then all passengers of class 3. 04:56.640 --> 05:02.230 Then ordered by the sixth column and alphabetically made it comes after a female. 05:02.240 --> 05:06.780 So we have a descending order starting with male and next. 05:06.840 --> 05:15.570 We as sort of a class the three males by the H also in descending order so starting from the oldest 05:15.570 --> 05:21.470 male until the youngest and let's have a check at the end of our data frame. 05:23.220 --> 05:25.760 So here we have the peasants class 1. 05:25.770 --> 05:34.360 We have females and ending actually at the youngest females here with the age to so the any end values 05:34.360 --> 05:35.910 are at the very end here. 05:35.920 --> 05:42.100 So to summarize we have sorted out our Titanic data frame first by the passenger class and then descending 05:42.100 --> 05:42.380 manner. 05:42.400 --> 05:45.090 Then by a and a descending man and then by H. 05:45.100 --> 05:46.180 And that is sending men. 05:46.810 --> 05:52.260 So what what if we want to sort our data frame again by the three columns here. 05:52.810 --> 05:56.730 But we want to start the passenger class in an ascending manner. 05:57.120 --> 06:01.000 And the sex in the descending manner and the H again in an ascending manner. 06:01.000 --> 06:06.340 And we can do this actually by not passing here at boolean value or false. 06:06.340 --> 06:12.100 True for all three columns but by passing here a list of boolean values. 06:12.250 --> 06:16.510 So let's assume we want to sort the peasants or class in an ascending manner. 06:16.510 --> 06:22.430 We have true then the sixth column in a descending manner falls. 06:22.700 --> 06:25.800 And then again the H column then an ascending manner. 06:25.830 --> 06:27.690 So also true. 06:27.740 --> 06:32.390 Let's see what we get here so now we are starting here. 06:32.390 --> 06:39.890 Passenger class from 1 soon an ascending manner then having all passengers of class 1 be Assad them 06:40.310 --> 06:42.460 by the gender starting from male. 06:43.570 --> 06:51.980 So we f here a descending order falls and then we are sliding all first class males by the H and then 06:52.070 --> 06:53.230 ascending manner again. 06:53.240 --> 07:01.010 So we feel truthful starting from the youngest male and let's again have a check here at the very last 07:01.030 --> 07:12.180 rows so here you have the peasants are classes 3 and we have females here and we would end actually 07:12.180 --> 07:16.270 at the oldest third class females. 07:16.320 --> 07:22.640 But now we have here at the end values at the very end but to summarize here so we canvassed for the 07:22.640 --> 07:29.720 site various method we can cite our data frame by a many columns and we can decide whether we want for 07:29.720 --> 07:34.510 each column in ascending or descending sliding all right. 07:34.510 --> 07:40.430 So now we're let's assume that we are happy with our sorting and then we can set our in-place parameter 07:40.430 --> 07:41.260 to a true 07:45.980 --> 07:49.250 and let's check in the first and the last rows here. 07:50.840 --> 07:59.030 So we have now 0 written our original data frame and we have our new data frame sorted by peasant class 07:59.690 --> 08:01.550 sex and age. 08:01.850 --> 08:07.910 And now let's assume that we are not happy with the sorting and we want to sort again our Titanic data 08:07.910 --> 08:10.030 frame by the index. 08:10.130 --> 08:16.330 So here we can see on the left side our data frame is starting with the index label 305 formed at forty 08:16.330 --> 08:25.840 five and so on and we can slide our data frame by the index label with the method thought index and 08:25.840 --> 08:29.700 again we have the option to have an ascending or descending manner. 08:29.710 --> 08:32.500 So we want to f our data frame sorted. 08:32.590 --> 08:41.160 Starting from 0 to 800 91 so we want to f in ascending order and we want to change our data frame back 08:41.160 --> 08:47.490 so we set to the in-place parameter to true and check again. 08:47.490 --> 08:53.460 So here we f again our it's no data frame you're starting with the peasant just with the index that 08:53.460 --> 08:56.390 Labor's 0 1 2 3 and so on. 08:56.380 --> 09:00.990 All right so now we are finished with this session and I hope to see you next on by.