WEBVTT 00:00.450 --> 00:06.660 Hello, everyone, in this session, we will discuss about sorting and merging of data frames, so the 00:06.660 --> 00:12.090 first thing which we will discuss is sorting data by one or more columns in ascending or descending 00:12.270 --> 00:12.700 order. 00:13.260 --> 00:18.510 Then we will discuss how we can combine data frames vertically or horizontally. 00:18.870 --> 00:24.640 Then we will learn about merging them, using keys left, right, in and out the joint. 00:25.050 --> 00:26.760 OK, so let's begin with it. 00:26.790 --> 00:30.890 So initially we will import Fondas and the number of. 00:35.760 --> 00:44.550 After importing these, I will be creating a new data frame, so this data frame is created from values. 00:46.500 --> 00:53.010 It has a shape of 20 cross food and it has values from two to eight. 00:53.430 --> 00:57.060 And the column names are A, B, C and the. 00:59.790 --> 01:06.300 This is the data frame which has been created, this data frame has 20 rows. 01:07.470 --> 01:16.560 And four columns now we will try to sort these using column E to sort these using column E, we will 01:16.560 --> 01:18.690 use D.F. dot sort values. 01:18.870 --> 01:26.820 So our values is a function which is used to sort of values in a Dufrene and we give the column by which 01:26.820 --> 01:33.340 we want to sort this out as the initial, the first parameter and the next barometer is in place. 01:33.930 --> 01:40.320 So in place operator here is means that if we give in place equal to do, then it will change the data 01:40.320 --> 01:41.130 frame itself. 01:41.580 --> 01:44.700 So if I change this. 01:46.020 --> 01:48.480 Then it will modify the data frame itself. 01:48.720 --> 01:54.270 So if I print the data frame, the data frame will be sorted with respect to it. 01:54.780 --> 02:00.960 Now, another thing, what I can do is I could have not given in basically due to if I did not give 02:00.960 --> 02:05.400 in physical truth, then the changes will not take effect in the device. 02:05.400 --> 02:11.640 So in that case, if I am not giving in basically true, then I would have to write it in a form like 02:11.640 --> 02:13.500 D.F. is equal to the sort. 02:13.830 --> 02:15.300 And without this. 02:17.270 --> 02:22.630 So if I am not writing in, please, then I have to give this particular format to make it effective, 02:22.970 --> 02:29.820 otherwise I can simply the idea of sort values, I'm simply saying, please, equal to two. 02:30.480 --> 02:37.840 Now, this is for sorting one column if you want to sort this column in descending order. 02:38.270 --> 02:41.850 And then what we can do is there is of the bottom of it, just ascending. 02:42.140 --> 02:45.440 So if by default ascending is said to be true. 02:46.390 --> 02:53.220 But in case we want to sorted in descending order, then we will have to write a sending people to Fuld's. 02:53.500 --> 02:59.320 So if you will put this ascending hold for us, then we saw this in descending order. 02:59.320 --> 03:03.040 So you can see that the column is decreasing. 03:03.580 --> 03:05.080 Now on a IS. 03:05.110 --> 03:09.310 We can also sort the values on the basis of multiple columns. 03:09.490 --> 03:15.850 So when we are sorting columns on the basis of multiple columns, in that case we will have to provide 03:16.000 --> 03:20.410 the argument of column which was given here. 03:20.800 --> 03:27.040 Initially we were giving just one column name here, but now we will have to give a list of column names. 03:27.220 --> 03:29.440 So we will give a list of column names. 03:29.710 --> 03:36.210 And based on the list of column names, we can give ascending values again in form of list. 03:36.640 --> 03:38.210 So you can see the difference here. 03:38.230 --> 03:40.180 Initially we were giving just. 03:41.590 --> 03:49.510 A Anderson default falls now we are giving with the values in a list for them, so once we do this, 03:50.020 --> 03:52.510 the values are sorted accordingly. 03:52.510 --> 03:55.270 So it has been sorted based on that. 03:55.510 --> 04:03.670 And on top of that, the values are sorted on the basis of the and then the values have been sorted 04:03.670 --> 04:11.320 on the basis of C, B in ascending order, the C in descending order of. 04:13.240 --> 04:22.180 Next thing, what we can do as of now, this source, the data by Colombe in ascending order but sorting 04:22.180 --> 04:24.350 by golems happens in descending order. 04:24.550 --> 04:33.280 So initially it was sort all the values of B in ascending order and then the values of B, that here 04:33.280 --> 04:35.950 we have three values as two. 04:36.250 --> 04:36.640 Right. 04:36.760 --> 04:46.630 So these values will be sorted for the on the basis of C, so it will force or B and after sorting B, 04:46.810 --> 04:53.160 the values corresponding to B will be sorted and these values will be sorted accordingly. 04:55.260 --> 05:04.920 Now we'll look at how we can combine different data set, so here I'm creating a data frame V1 and this 05:04.920 --> 05:10.320 data frame consists of volumes of letters and numbers. 05:10.330 --> 05:14.250 So there are letters and the numbers, one and two. 05:14.610 --> 05:16.690 And I've given the column names also. 05:17.550 --> 05:21.150 Similarly, I have created another data frame with value. 05:21.180 --> 05:23.070 See these three four. 05:23.340 --> 05:26.150 Gandal So I just run this now. 05:28.120 --> 05:33.640 So these are the values which has been created, these are the things which have been created now, 05:34.360 --> 05:41.520 as you always remember, those are equivalent to Zettl and columns are always equivalent to one. 05:41.770 --> 05:43.510 So we always did this. 05:43.510 --> 05:46.990 We just remembered using this rose. 05:48.150 --> 05:57.990 Former columns, and this is written in a way zero one one, so whenever I see Zettl, it means we are 05:57.990 --> 06:02.640 working with rules and whenever I see one, it means we are working with columns. 06:06.880 --> 06:14.790 So here what we are doing is we are concatenating, we are combining, we are just fisting them together. 06:14.920 --> 06:22.120 So we are basting our data from one data frame two together and we are basting them on the basis of 06:22.120 --> 06:26.370 access to this means that we are facing them always. 06:26.530 --> 06:29.380 So that means the rules of the event. 06:29.380 --> 06:36.190 One will be present at the top and after that the rules of data frame will be present. 06:36.880 --> 06:42.350 Now, whenever the values for the data frame one are missing. 06:42.580 --> 06:47.400 So as you see in data frame one, we do not have the column animals present. 06:47.830 --> 06:49.210 So what happens is. 06:50.160 --> 06:53.910 Whatever the value is not present, it gives not a number there. 06:54.690 --> 06:59.300 So here we have concatenated these on the basis of X is zero. 07:00.250 --> 07:06.950 OK, so in case more columns were present in data frame to all those columns will be added here and 07:06.980 --> 07:09.200 the values will be given as not a number. 07:09.900 --> 07:16.730 Now, similarly, if we are doing the same thing now, if you one more thing to notice here is the indexes 07:16.730 --> 07:18.680 have we have not been modified. 07:18.890 --> 07:22.340 So in my first data frame, the index was zero one. 07:22.340 --> 07:24.070 Here again, the index was zero one. 07:24.080 --> 07:29.320 So while sticking them together, it has actually retained the original indexing. 07:29.600 --> 07:35.180 So if you want to make the indexing again, correct the index, then what we can do is we can simply 07:35.390 --> 07:38.000 ignore index and mean said that through. 07:38.270 --> 07:39.920 So when we said this thing to. 07:41.260 --> 07:42.430 What happens is. 07:43.660 --> 07:45.780 Let me just run this again. 07:47.080 --> 07:55.660 So when we put this concatenated this, so what happens is it will simply retain the create a new index 07:55.660 --> 07:59.680 for this so it will change the indexes from zero one zero one two zero one, two, three. 08:00.340 --> 08:05.840 Now, another thing which we can try here is I'm creating a new data frame, data frame three. 08:06.070 --> 08:14.260 Now, this data frame consists of a column Unimin and another column name, and it has three rows of 08:14.260 --> 08:17.940 data, which is board volume monkies, your Staiger John. 08:18.160 --> 08:21.050 So these are the values which are present in this particular detainee. 08:22.910 --> 08:27.030 So we are trying to concatenate these by axis one. 08:27.290 --> 08:31.510 Now, what this means is that we will have this particular data frame. 08:32.030 --> 08:33.800 We will have the data frame to. 08:35.380 --> 08:41.050 This particular data frame, this is the data pointing to a different data frame to heal forced. 08:43.770 --> 08:46.350 So this is day two, sorry. 08:53.660 --> 08:59.410 This is the data frame do this is to be the frame to be, and what we're trying to do is we are going 08:59.420 --> 09:01.760 to be needing data from two would be to frame three. 09:01.910 --> 09:08.540 Now, what happens is the full data frame has two rules and three columns. 09:09.020 --> 09:12.970 The second data frame has three rows and two columns. 09:13.190 --> 09:18.440 So as a result, then we get five columns. 09:18.650 --> 09:22.040 That is letter number, animal animal. 09:22.250 --> 09:28.790 I mean, this is what we get by concatenate them with access one, which is column based concatenation. 09:29.150 --> 09:34.690 And what we get is we again get because the rule number is two here and three here. 09:34.700 --> 09:36.380 So it creates three rules. 09:36.620 --> 09:45.370 And wherever we don't have values from this particular data frame, it makes them as not a number. 09:45.530 --> 09:48.170 So it improves the value as not a number. 09:48.350 --> 09:54.980 And once we have done that, once we have done all the sorting, once we have done the combination, 09:55.490 --> 10:00.770 whatever, merging, whatever we are trying to do, what we can do is we can then improve the values, 10:00.890 --> 10:02.450 how we view the values. 10:02.450 --> 10:03.560 We see that little. 10:03.710 --> 10:10.000 But for now, just try to understand that these values will get generated and it will not have an empty 10:10.070 --> 10:10.340 value. 10:10.340 --> 10:11.270 It will have more than that. 10:12.470 --> 10:14.720 Now, the next thing what we can try is. 10:16.100 --> 10:21.620 So we Maskin got donated by Accies one, similarly, we are trying to concatenated Dufrene one, we 10:21.630 --> 10:22.490 did frame three. 10:22.640 --> 10:27.350 So what it does is it again creates the values, not no matter what. 10:27.350 --> 10:28.610 The values were not present. 10:32.740 --> 10:40.390 Now, let us discuss how we can move to different data frames, so here I am creating other data from 10:40.390 --> 10:41.410 the data frame. 10:41.410 --> 10:48.430 One contains one column customer data, which contains volumes one to five, then out of the column 10:48.430 --> 10:49.570 contains the product. 10:50.140 --> 10:54.880 Then that is another data frame which contains the customer I.D.. 10:56.180 --> 10:57.920 And along with the customer itis. 10:59.400 --> 11:01.800 It contains the state now. 11:01.970 --> 11:03.960 Now, let me run this. 11:05.300 --> 11:12.800 So this is the data frame when this is the data frame, too, so here we have customarily and you can 11:12.800 --> 11:15.890 see that for the similarity, one, two, three, four, five. 11:15.900 --> 11:17.120 These are unique values. 11:17.420 --> 11:20.480 And we have video footage which are kind of repeating. 11:21.320 --> 11:26.000 Then here we have another the customer I.D., which are different from the customer I.D., which are 11:26.000 --> 11:27.230 present in this data. 11:27.260 --> 11:27.770 Name one. 11:29.920 --> 11:38.890 So we are next trying to merge these data frames, so why we are merging the data frames, what we are 11:38.890 --> 11:43.570 doing is we are merging the data from one with data frame two. 11:44.930 --> 11:52.460 And we are merging it on customer already, so we are seeing feed thought much and we say if one the 11:52.640 --> 11:59.930 to, then we give the key on which we want to merge these ideas and how we want to merge this. 12:00.170 --> 12:02.810 So this is the type of joint which we are trying to have. 12:03.140 --> 12:04.600 So this is in our joint. 12:05.360 --> 12:14.780 So when we say in our joint, that means that we want to keep only the rules which are common between 12:14.780 --> 12:17.390 this Dufrene one and the same two. 12:17.750 --> 12:22.010 And we want to keep the commonness on the basis of the customer. 12:22.700 --> 12:24.830 So we have customer varieties. 12:24.830 --> 12:29.140 One, two, three, four, five year and we have three, four, five, six, seven. 12:29.390 --> 12:33.190 So the common customer IDs are three, four and five. 12:33.440 --> 12:39.080 So we are expecting to have only three rules in this particular little thing. 12:39.110 --> 12:42.090 So while I run this, I get three rules from the state of Maine. 12:42.500 --> 12:50.660 Now, the next is we want to join on customer idea again, but we are wanting to join it on the outer 12:50.660 --> 12:50.970 join. 12:51.140 --> 12:58.520 So what our vision gives us, it gives us rules from all of the things, all the rules from the date 12:58.520 --> 12:58.940 of names. 12:59.120 --> 13:00.170 So we will get. 13:01.330 --> 13:07.730 Customer added one, two, seven, all of these customer IDs would be present. 13:08.020 --> 13:14.560 So once we've done this, we get customer I.D. one to seven and all the details in front of them now 13:14.560 --> 13:20.230 because customer IDs, six and seven were missing in the frame one. 13:20.380 --> 13:23.830 So the product values have been imputed as not the number here. 13:24.190 --> 13:28.180 And because the customer IDs, one and two are missing in between two. 13:28.390 --> 13:31.830 So that is the reason why they see this coming as not the number here. 13:33.120 --> 13:37.390 Now, next is customarily merging on customarily an. 13:38.640 --> 13:43.740 I joined this group, so I left join will basically give us. 13:45.060 --> 13:50.110 These are the two data frames which we have and we are keeping data from one on the left hand side. 13:50.340 --> 13:54.990 So this way we will get the columns which are from the. 13:56.950 --> 14:03.280 We will get the rules which are present in the next day, the family, so the rules are one to five. 14:04.430 --> 14:10.660 So when we are doing left joined, we are getting close one to five and the details which are missing 14:10.670 --> 14:16.010 on the right side, thebus that has been imputed, that has been created. 14:16.580 --> 14:23.650 Now, similarly, when we have this right joint, so the columns of the rules are picked from the right 14:23.660 --> 14:27.700 inside the ring, which is date of name to which has rules three to seven. 14:27.710 --> 14:29.060 So all of these are present. 14:29.330 --> 14:35.720 And the product, the product, the definition does not have the rule six and seven. 14:35.930 --> 14:38.450 So that is why the number has been created here. 14:39.520 --> 14:42.830 Now, a similar kind of things you can try to own here. 14:42.850 --> 14:50.530 We have more combinations so you can see we have the frame 20 frame three in the frame four. 14:50.740 --> 14:56.990 So now here what we have is in the frame three, we have values repeated. 14:57.940 --> 15:01.120 So the date of the T has to twice. 15:02.090 --> 15:03.830 And it has a lot on unbottled. 15:04.830 --> 15:12.760 While the Dufrene Ford has won the bidding twice and it doesn't have an number here. 15:13.530 --> 15:15.000 So what we do with. 15:17.250 --> 15:25.070 So now when we merge these different three days of rainfall, one customer arrive and have inner join, 15:25.380 --> 15:33.510 then what we get as a result is because the Zatarain three ended up rainfall has customer already one. 15:36.760 --> 15:37.330 For. 15:38.600 --> 15:47.060 In phone, so we get one, two and four, and the repetitions with the president are coming as it is 15:47.420 --> 15:54.230 so because the best humanity has repetition in the deafening for so it is taking what the values and 15:54.620 --> 15:55.940 values on both of these. 15:56.210 --> 16:00.280 And because the cost when I was presenting the different three twice. 16:00.470 --> 16:04.430 So it is generating values on top of these and facing the values accordingly. 16:05.240 --> 16:11.130 Similarly, all the violence will be imputed when we are generating a left join order, I join. 16:11.480 --> 16:14.490 So please try all these scenarios. 16:15.260 --> 16:15.770 Thank you.