WEBVTT 00:01.440 --> 00:08.700 Now, let us go ahead and give you different details about the date of this, so let me have this data 00:08.700 --> 00:15.820 from the witness age of my fiancee, this education before violence, all of these problems present. 00:16.260 --> 00:20.640 So the first thing what we can do is we can check the different type of problems which has. 00:20.910 --> 00:24.530 So we can say the problem is. 00:26.720 --> 00:33.650 This gives us the list of different problems that have arisen and an important function is D.F. Thought 00:33.700 --> 00:38.180 Info, which gives us top level information about the. 00:40.620 --> 00:51.240 So here we can see that there are four thousand five hundred ten rows of data and. 00:52.470 --> 00:55.140 These are different from. 00:56.430 --> 01:01.290 In full, this is the age. 01:04.300 --> 01:05.560 Then this is. 01:06.830 --> 01:18.950 The job is the marital status, the education level, the default, the violence, housing details, 01:18.950 --> 01:24.570 loans, deals, all the biggies, these other details related to it, other things. 01:24.620 --> 01:26.120 What we can check is. 01:32.520 --> 01:40.200 Is the describe, which gives us a summary of the numerical problems, so out of the numerical column, 01:40.200 --> 01:47.510 we can see that this is the each column which has counts as forty five thousand two hundred and eleven. 01:47.760 --> 01:55.830 The mean age is is going to be the standard deviation is often the minimum age is 18, the maximum age 01:55.830 --> 01:56.960 is ninety five. 01:57.360 --> 02:04.680 And you can see that the maximum value, which is seventy five percent, has a value as what, eight, 02:04.920 --> 02:10.580 which means that there is a huge gap between the seventy five percent and the maximum value. 02:10.890 --> 02:16.290 So we can consider getting rid of the outlier values in the H. 02:17.530 --> 02:25.120 So these are a few insights which we can get from these volumes and bring out those structures, apart 02:25.120 --> 02:30.970 from that, we can take different vibes of the volume so we can take the detailed. 02:34.240 --> 02:42.130 This gives us the details of the volumes so we can see that each is individual job is object here, 02:42.130 --> 02:50.140 these object data VIPs means that these are strings and you can see that we have these values. 02:50.560 --> 02:57.130 So other values we can check if a particular value is numeric or North American needs and or take care 02:57.130 --> 02:57.670 of that. 03:00.530 --> 03:08.600 So here it seems that all the values at that, but in case if we have a column that should actually 03:08.600 --> 03:15.290 be numeric in nature and is in strength, then we can change the date of the column itself. 03:15.770 --> 03:17.870 So these are different things which we can do. 03:19.320 --> 03:24.270 Then we can check the shape of the data frame using the cheap. 03:26.910 --> 03:29.100 So this is the shape of the data frame. 03:29.110 --> 03:32.610 There are 17 volumes and these many number of those. 03:33.750 --> 03:41.220 Now, if you want to filter out some data from this particular data frame, then there are several ways 03:41.220 --> 03:41.940 of doing it. 03:42.210 --> 03:46.210 One is using a location so we can see if. 03:47.530 --> 03:57.220 Got a lock, and inside this, we can give the location values, so let us see, I want to view of the 03:57.220 --> 04:04.510 first row so we one and I want to see the second column. 04:06.590 --> 04:08.270 So it has the value of single. 04:10.470 --> 04:15.690 We can see the same thing using the left hand. 04:18.640 --> 04:21.480 So you can see this value is single now let us see. 04:21.520 --> 04:29.140 We want to have a sequence of values so I can see it from let's say I want to see the fourth row and 04:29.140 --> 04:31.660 I want to see the values in the fifth column. 04:33.430 --> 04:34.850 So I can do it this way. 04:35.320 --> 04:38.870 So using I is not really suggested. 04:39.070 --> 04:47.770 So instead of using I love, we actually use the location function because I love me because of having 04:48.490 --> 04:52.780 this dependency and confusion while working the data frames. 04:52.960 --> 04:59.250 So instead of using I love to actually work with different Farlam names itself. 05:00.370 --> 05:02.050 Let us try to filter out. 05:03.030 --> 05:08.220 So for filtering out these, we can use. 05:09.470 --> 05:20.360 There and then we give the column name usually, so the week now this gives us the column now later 05:20.360 --> 05:25.070 said they don't want to get only one column, but multiple columns. 05:25.280 --> 05:27.500 In that case, we will have to pass Mr.. 05:32.550 --> 05:43.470 So we have the next 12 minutes on this show, then the next column, which I want to do is. 05:47.380 --> 05:51.200 So I can get these quotes now. 05:51.330 --> 05:57.060 I think what we can do is see we don't want these columns. 05:57.080 --> 06:04.330 I don't want all your details of data, but we want to put these out on the basis of a particular condition. 06:04.930 --> 06:09.620 So in that case, what we can do is we will have to use clock. 06:10.000 --> 06:11.200 So what we do is. 06:12.910 --> 06:17.290 He can simply see the dot look. 06:18.490 --> 06:24.530 Inside this, they have to be, which we are fudging. 06:25.120 --> 06:27.320 So we can give the condition. 06:27.800 --> 06:33.130 So let's say we want to get the condition as see the. 06:36.540 --> 06:37.110 Now. 06:38.060 --> 06:43.890 We have to give the conditions in the down bracket and then we will in one condition at every point. 06:44.090 --> 06:45.910 Now, what is the condition which I want to have? 06:45.920 --> 06:46.510 Let us see. 06:46.520 --> 06:50.580 I want to have age where the age of the person is more than 75. 06:51.080 --> 06:56.120 OK, so for that I can simply see D.F.. 06:58.570 --> 07:02.350 I'm against cloning, so I h. 07:03.380 --> 07:08.460 I want people I want this each week, seventy 75. 07:09.450 --> 07:18.750 Now, on this, I get on the data and the INS has indicated that 70 five now legacy, I want the but 07:18.780 --> 07:22.950 also being so in that case, I will give under the conditions. 07:22.950 --> 07:26.580 So let us say I want on condition you OK? 07:26.790 --> 07:34.620 And I don't want to give under the condition which is C. having education as primary. 07:36.060 --> 07:37.550 So I would say the. 07:38.660 --> 07:40.490 Hi, the name. 07:41.990 --> 07:42.980 Education. 07:44.740 --> 07:47.830 And this education value has to be private. 07:48.220 --> 07:50.440 So I say, is he to. 07:51.640 --> 07:52.000 Brian. 07:53.820 --> 07:54.960 OK, so. 07:56.550 --> 07:59.520 Like this, I can get the details that age is good enough. 07:59.550 --> 08:01.690 Seventy five, education is primary. 08:01.980 --> 08:02.790 I'm like this. 08:02.790 --> 08:04.950 I can keep on adding different conditions. 08:05.190 --> 08:10.160 Now, once I'm adding condition, legacy, I want to have these conditions also. 08:10.350 --> 08:14.910 And I also want to I don't know the columns which I want to have. 08:15.150 --> 08:19.170 So what I can do is just against these conditions. 08:19.380 --> 08:26.550 So I have these given these conditions now after these conditions, I just give a comma and after all 08:26.610 --> 08:29.850 my all, which I wouldn't have. 08:29.850 --> 08:34.130 So I have an agent of education. 08:34.140 --> 08:37.500 So let me give each an education so I can give each. 08:38.500 --> 08:43.330 They give education and let me give. 08:46.080 --> 08:53.310 So I can get it out and I can get more education based on the criteria which I have, if I. 08:56.300 --> 09:04.550 Now, let's see, we want to have these conditions with just fine now in case we want to have any negative 09:04.550 --> 09:05.140 condition. 09:05.450 --> 09:10.540 So in that case, we can simply put all negation sign in front of this. 09:11.420 --> 09:17.390 So just to get in front of this so it will give everything apart from this particular condition. 09:18.390 --> 09:24.900 Now, other thing, what we can do is let's see if we want to drop a particular column, so in case 09:24.900 --> 09:30.690 we want to drop a particular column, then what we can do is we can save the ABC 09:33.180 --> 09:34.930 and save this function. 09:35.190 --> 09:38.330 They can give the list of columns which we want to. 09:38.940 --> 09:43.250 So I get to look at the list of each of. 09:45.630 --> 09:47.790 So I give you the quote. 09:49.850 --> 10:05.040 Often not the axis, which I want to get to this, so I have easy access to one, this is what you want. 10:05.080 --> 10:06.440 So I can simply say this. 10:08.380 --> 10:14.010 So once I have this, it will remove those two columns from the plane. 10:14.380 --> 10:23.190 Now, one thing you notice here is if we again run now, if we again run this war and the. 10:25.330 --> 10:25.870 So. 10:27.290 --> 10:35.450 If you check, the head still has the goal of open while it has been for you, so it will actually make 10:35.450 --> 10:40.610 this change, what we have to do is if you are going in this way, then we will have the same 10:43.370 --> 10:48.410 sort of thing over and over again this way. 10:48.420 --> 10:54.420 Then it will just show you what it will be very different. 10:54.470 --> 10:56.960 So this is one that will be liberal. 10:57.520 --> 10:59.690 And another way we will be the this. 11:01.150 --> 11:03.770 That we can write in the equal to. 11:04.860 --> 11:12.170 So what we can do is we can simply type of in place equal to do in front of it. 11:13.240 --> 11:15.420 So we can simply say Obama. 11:16.620 --> 11:28.530 In equal to then I think from this now out of the way of doing this is we can simply delete. 11:29.480 --> 11:35.780 So what we can do is we can simply see then and then we can give the name. 11:39.870 --> 11:48.120 I'm there and then we can give the qualifying age, so then it will delete the goal of age from this. 11:48.840 --> 11:51.480 So if we see the again. 11:55.620 --> 12:02.820 So now the D.F. will not have the goal of each and another thing, what we can do is we can delete the 12:02.820 --> 12:05.040 entire data frame by then. 12:05.670 --> 12:10.470 I'm just simply saying I like this. 12:11.280 --> 12:14.460 So these are a few basic operations which we have. 12:14.790 --> 12:19.200 Apart from this, there are several other operations we would have to do. 12:19.530 --> 12:27.480 These are operations which will allow us to basically analyze the data initially and view the detail 12:27.510 --> 12:28.010 properly. 12:28.310 --> 12:34.550 But we will have to do a lot more on top of these on top of these dolphins. 12:34.740 --> 12:43.940 So we will see the entire process, which we will be following for data transformation and data cleaning. 12:44.160 --> 12:46.400 So let us go to that. 12:46.440 --> 12:52.440 I'm going to see this in the next session, how we will work with actual data and what are the things 12:52.440 --> 12:53.740 you need to take care of? 12:53.760 --> 12:58.560 What are the things you need to consider while analyzing and modifying your data? 12:58.620 --> 13:01.130 That is something we will look after this session.