Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated: 1 00:00:00,750 --> 00:00:02,320 Hello everyone and welcome to this. 2 00:00:02,340 --> 00:00:08,100 Part 2 of the overview of data operations lecture and this lecture we're going to continue our understanding 3 00:00:08,390 --> 00:00:11,030 of some of the most common data frame operations. 4 00:00:11,040 --> 00:00:13,810 Let's go ahead and jump to our studio and get started. 5 00:00:14,310 --> 00:00:19,620 All right here your studio right where we left off the last lecture talking not referencing columns 6 00:00:19,620 --> 00:00:19,860 . 7 00:00:19,890 --> 00:00:24,390 We're going to continue talking about data from operations by talking about adding rows to the data 8 00:00:24,390 --> 00:00:25,350 frame. 9 00:00:25,350 --> 00:00:30,900 Let's go ahead and clear this data frame for the consul excuse me and talk about topic number seven 10 00:00:31,020 --> 00:00:43,660 adding rose to go ahead and create a data frame ZF to and call data frame pass in a column name call 11 00:00:43,670 --> 00:00:56,640 the name that one set of equal to two thousand and say call by named to and equal to the string or character 12 00:00:56,730 --> 00:00:57,250 you. 13 00:00:57,270 --> 00:01:03,660 So if I take a look at this DFA to notice has the same column names as the data from or working before 14 00:01:03,970 --> 00:01:05,120 the earth. 15 00:01:05,220 --> 00:01:07,070 Except now I have two new entries. 16 00:01:07,200 --> 00:01:13,260 The value 2000 and new in order to bind this new roads for a data frame. 17 00:01:13,400 --> 00:01:17,880 All we have to do is use the our bind function that we already used in the past. 18 00:01:17,880 --> 00:01:26,790 So I can say DFW call our Bines for binds and then I'm going to go ahead and pass on my original data 19 00:01:26,790 --> 00:01:33,890 frame D.S. and then DFI to which is that data frame that I want to bind as a new row. 20 00:01:34,080 --> 00:01:38,690 And now if I take a look at DFW you'll notice down at the bottom we have a new row. 21 00:01:38,730 --> 00:01:42,380 New value is 2000 and you. 22 00:01:42,390 --> 00:01:43,650 Now let's shift our focus. 23 00:01:43,650 --> 00:01:46,650 Talking about adding new columns to a data frame. 24 00:01:46,800 --> 00:01:48,770 There's a couple of different ways to do this. 25 00:01:48,770 --> 00:01:50,820 I'm going to go ahead and show you a few of them. 26 00:01:51,040 --> 00:01:58,470 Going to go out and clear the console and color ADF which is the original data frame column name one 27 00:01:58,470 --> 00:02:01,490 column name two rows one through 10. 28 00:02:01,860 --> 00:02:08,060 I can add a new column using the dollar sign method which is a dollar sign. 29 00:02:08,370 --> 00:02:14,660 Create a new column name and then pass in whatever you want the column to be. 30 00:02:14,850 --> 00:02:21,120 So imagine I wanted a new column that was just double values of column name one where I can go ahead 31 00:02:21,120 --> 00:02:30,760 and do say two times and then pass in the data frame if dollar sign called out Name one. 32 00:02:30,960 --> 00:02:38,440 And now if we look at B-F you'll notice we have a value new call for this new column and it's just double 33 00:02:38,440 --> 00:02:40,390 the value of column name 1. 34 00:02:40,810 --> 00:02:46,500 And that's one way you can quickly create new columns onto a data frame instead of using something like 35 00:02:46,950 --> 00:02:49,090 row binder see buying that we saw earlier. 36 00:02:49,200 --> 00:02:54,210 It's much easier to just go ahead and name your new column directly with some sort of assignments of 37 00:02:54,210 --> 00:02:55,310 that new column. 38 00:02:55,320 --> 00:03:01,830 Keep in mind that these values should line up as far as a number of elements for your column. 39 00:03:01,830 --> 00:03:05,510 This sort of operation is also really useful if you want to make copies of columns. 40 00:03:05,520 --> 00:03:11,550 So for instance we take a look at if we have called Name one called Name 2 and in that new column let's 41 00:03:11,550 --> 00:03:19,500 say I wanted to make a copy of my new column I could say if dollar sign new call and have that just 42 00:03:19,500 --> 00:03:23,740 be equal to Deif dollar sign you call. 43 00:03:24,090 --> 00:03:30,630 But instead of calling it you call it I'm going to do is say you call that copy. 44 00:03:30,630 --> 00:03:33,720 Now if I do this it's going to go ahead and check out the name. 45 00:03:33,720 --> 00:03:34,940 The head of media. 46 00:03:35,430 --> 00:03:38,600 You'll see I have a copy of that new column. 47 00:03:38,610 --> 00:03:42,250 You can also use any other of the column references we've talked about. 48 00:03:42,260 --> 00:03:57,240 So for instance I could say D.S. brackets comma and then put in new call let's say copy to then assign 49 00:03:57,240 --> 00:04:02,180 this to DMF new call. 50 00:04:02,190 --> 00:04:06,570 So now if I check head of D.S. I have the second copy here. 51 00:04:06,630 --> 00:04:10,870 So I have new call you call the copy new call the copy to. 52 00:04:10,890 --> 00:04:16,620 The only difference between this sort of operation and this line that I'm highlighting versus this operation 53 00:04:16,740 --> 00:04:18,860 is just the way I'm addressing the column. 54 00:04:18,960 --> 00:04:25,410 So classic notation is just dollar sign method but you can also use the brackets and comma method for 55 00:04:25,440 --> 00:04:26,940 denoting that new column. 56 00:04:26,970 --> 00:04:29,130 It's really up to you and what you feel more comfortable with. 57 00:04:29,280 --> 00:04:35,100 But the basic premise is that you call a column as if it's already on your data frame but remember to 58 00:04:35,100 --> 00:04:40,800 give it some new assignment for new values and that's the basic way of adding new columns to your data 59 00:04:40,900 --> 00:04:41,340 . 60 00:04:41,640 --> 00:04:44,550 So we've gone over adding rows adding columns. 61 00:04:44,550 --> 00:04:47,400 Up next is setting column names. 62 00:04:47,400 --> 00:04:53,940 We already know and went over that if we just say call names we can get back the names of the columns 63 00:04:53,940 --> 00:05:01,600 of our data frame if we want to actually rename columns what we can do is use column names. 64 00:05:01,620 --> 00:05:08,140 Passen are data frame and there's two things we can do if we wanted to rename all of them at once. 65 00:05:08,190 --> 00:05:14,940 We could just pass a vector of new names so I could say one and that isn't just passing characters. 66 00:05:14,940 --> 00:05:21,180 These aren't actually integers 3 for what's good and say five. 67 00:05:21,180 --> 00:05:28,460 So I have columns 1 2 3 4 5 so we'll rename them to just those numbers 1 2 3 4 5 and then if we check 68 00:05:28,460 --> 00:05:35,000 out the head of our data frame notice now instead of those original column names you just have 1 2 3 69 00:05:35,000 --> 00:05:36,310 4 and 5. 70 00:05:36,560 --> 00:05:41,330 So that's what you could do if you wanted to rename all the columns at once if you just wanted to rename 71 00:05:41,330 --> 00:05:48,830 a single column but you'd end up doing is calling call names DPF and then with brackets you would go 72 00:05:48,830 --> 00:05:52,130 ahead and select what column number you want to rename. 73 00:05:52,130 --> 00:05:54,430 So for instance let's say I want to rename the first column. 74 00:05:54,560 --> 00:06:00,380 I'll say one and then I'll just say new column name. 75 00:06:00,440 --> 00:06:02,400 Pass that as a string. 76 00:06:02,420 --> 00:06:08,500 And if I check the head of the IDF notice I have a new code name for that first column and my data frame 77 00:06:08,500 --> 00:06:08,790 . 78 00:06:09,080 --> 00:06:13,910 And again quick note here this is an integer not a character. 79 00:06:13,960 --> 00:06:14,380 All right. 80 00:06:14,420 --> 00:06:19,210 So that's the basic overview of the topic of column names in a data frame as far as setting them. 81 00:06:19,540 --> 00:06:24,320 You have a few more topics go over and those are selecting multiple rows selecting multiple columns 82 00:06:24,410 --> 00:06:26,150 and then dealing with missing data. 83 00:06:26,150 --> 00:06:27,930 Let's talk about how to select most Choros. 84 00:06:27,950 --> 00:06:32,460 We've already gone in the first part of this lecture series on how to select the single row. 85 00:06:32,600 --> 00:06:35,710 Let's talk about selecting multiple rows. 86 00:06:36,020 --> 00:06:36,960 Actually quite easy. 87 00:06:37,010 --> 00:06:40,520 And it's really similar to how we selected the multiple rows in a matrix. 88 00:06:40,550 --> 00:06:46,460 All you have to do is put in the name of your data frame and then what you can do is go ahead and slice 89 00:06:46,580 --> 00:06:47,330 the rows you want. 90 00:06:47,330 --> 00:06:53,610 So if you wanted the first 10 rows you would just use or slicing notation comma. 91 00:06:53,750 --> 00:06:56,810 And that's one way of selecting the first 10 rows. 92 00:06:56,810 --> 00:06:59,900 You can go ahead and say selects first let's say three rows. 93 00:07:00,110 --> 00:07:02,080 Now returns first three rows. 94 00:07:02,090 --> 00:07:07,700 This is essentially the same as calling the head of your data frame both for a specific number of rows 95 00:07:07,850 --> 00:07:09,340 that you want returns. 96 00:07:09,360 --> 00:07:14,270 Now the way you can do this is just by using head sooner that if we just say head data frame. 97 00:07:14,320 --> 00:07:20,100 I'll return the first six rows but you can always specify how many rows you want back as far as these 98 00:07:20,120 --> 00:07:23,480 top rows and you can specify that a second argument. 99 00:07:23,480 --> 00:07:29,930 So let's say I want the first seven rows of my data frame I can say DF comma seven and I'll return the 100 00:07:29,930 --> 00:07:33,620 first seven rows. 101 00:07:33,620 --> 00:07:39,560 You can also take advantage of negative sign to select everything but a certain row. 102 00:07:39,590 --> 00:07:43,110 So imagine I wanted to select everything but row 2. 103 00:07:43,250 --> 00:07:45,290 I'm going to go out and clear the consul here. 104 00:07:45,560 --> 00:07:48,020 So we have data frame. 105 00:07:48,020 --> 00:07:52,280 Let's just say head of data frame and I want to select everything. 106 00:07:52,280 --> 00:07:53,770 But the second row. 107 00:07:53,810 --> 00:07:57,800 So notice how second row has a bunch of other unwieldy numbers there. 108 00:07:57,800 --> 00:07:58,840 We want to select everything. 109 00:07:58,850 --> 00:08:05,840 But that second row I can say with brackets we say to comma that would have selected the second row 110 00:08:06,260 --> 00:08:11,570 by say negative to comma that selects everything but the second row. 111 00:08:12,050 --> 00:08:12,550 OK. 112 00:08:12,550 --> 00:08:19,440 And so you can use those data signs in a few other operations with our in order to say everything. 113 00:08:19,490 --> 00:08:23,960 But now we'll go over those as they come up throughout the course. 114 00:08:23,960 --> 00:08:29,750 Finally I want to go over conditional selection and in order to do this we're going to be using empty 115 00:08:29,750 --> 00:08:33,240 cars data frame so that's empty cars. 116 00:08:33,410 --> 00:08:39,550 And it looks like this we can do conditional selection on a data frame by passing in logical conditions 117 00:08:39,560 --> 00:08:39,620 . 118 00:08:39,650 --> 00:08:43,190 We want to filter by and the syntax for that is as follows. 119 00:08:43,190 --> 00:08:48,290 I'll say something like empty cars brackets. 120 00:08:48,620 --> 00:08:56,840 And then let's say I wanted to pass or get back every car or every row where the MPG was greater than 121 00:08:57,020 --> 00:08:58,450 20 mpg. 122 00:08:58,610 --> 00:09:02,800 What I would then end up doing is assin the name of my data frame in this case. 123 00:09:02,810 --> 00:09:05,460 Empty cars dollar sign. 124 00:09:05,630 --> 00:09:10,070 MPG is the column I'm interested in and I will say greater then 20. 125 00:09:10,070 --> 00:09:12,810 So he used this comparison operator. 126 00:09:12,930 --> 00:09:13,520 All right. 127 00:09:13,520 --> 00:09:17,470 So if I just do this I'll get back in air. 128 00:09:17,550 --> 00:09:22,640 And the reason I'm getting back the air is because I have undefine columns selected. 129 00:09:22,640 --> 00:09:26,390 So what I actually need to do is remember the pass in it comma here. 130 00:09:26,600 --> 00:09:31,740 So students sometimes forget to pass and that comma and they get this undefine column selected. 131 00:09:31,940 --> 00:09:36,380 So remember that if you're getting this undefine column selected it's because you forgot to say Oh give 132 00:09:36,380 --> 00:09:38,630 me back all the columns for that. 133 00:09:39,410 --> 00:09:41,280 And now we get back the state of frame. 134 00:09:41,360 --> 00:09:43,430 And we can break this down by thinking. 135 00:09:43,790 --> 00:09:50,300 We're asking are for empty cars data frame where this statement is true for the rows where the column 136 00:09:50,300 --> 00:09:52,200 of them e.g. is greater than 20. 137 00:09:52,250 --> 00:09:55,310 So we're just saying give me back the rows where this is true. 138 00:09:55,640 --> 00:09:58,750 Comma all the columns for that. 139 00:09:58,760 --> 00:10:06,170 So then you can also pass in addition arguments over here for specific columns back. 140 00:10:06,170 --> 00:10:12,300 Let's go ahead and build on top of this example by filtering by two separate columns. 141 00:10:12,320 --> 00:10:16,190 So for example I have empty cars. 142 00:10:16,190 --> 00:10:25,950 Let's go ahead and show the head them see cars so mpg cylinders horsepower etc. I can say empty cars 143 00:10:27,490 --> 00:10:33,430 rackets I can put it in one condition like we saw earlier where I say empty cars. 144 00:10:33,540 --> 00:10:36,020 MPG is greater than 20. 145 00:10:36,280 --> 00:10:40,980 Well let's say I also wanted the number of cylinders to be equal to six cylinders. 146 00:10:40,990 --> 00:10:44,830 So that's this second column c y l cylinders. 147 00:10:44,880 --> 00:10:56,610 I would then say and put that and operator there and say and empty cars dollar sign c y l is equal to 148 00:10:56,670 --> 00:10:57,970 6. 149 00:10:58,170 --> 00:11:05,400 And remember to put in a comma here and sometimes it's also nice to put in parentheses around your logical 150 00:11:05,860 --> 00:11:11,790 for comparison operator statements just so it's a little easier to read you can see what are the two 151 00:11:11,790 --> 00:11:18,270 separate factors that you're trying to compare and you'll see we'll get back empty cars where the MPG 152 00:11:18,570 --> 00:11:25,260 was greater than 20 and the number of cylinders was equal to six if you wanted to only get specific 153 00:11:25,260 --> 00:11:26,790 cones back from this. 154 00:11:26,790 --> 00:11:29,820 You could add that in as a second argument over here. 155 00:11:29,820 --> 00:11:35,320 So for instance let's say we only wanted to get HP or horsepower back for these cars. 156 00:11:35,650 --> 00:11:48,130 Well it's a MPG cylinder and HP I couldn't pasan a vector of those column names say mpg cylinder horsepower 157 00:11:48,120 --> 00:11:48,970 . 158 00:11:49,620 --> 00:11:51,990 And now we only get back those columns. 159 00:11:52,000 --> 00:11:56,430 This is the kind of filtering you're going to be doing all the time when you're importing data from 160 00:11:56,430 --> 00:11:59,430 a CSFB and playing around a fit trying to visualize it. 161 00:11:59,620 --> 00:12:01,150 Get an idea of what it looks like. 162 00:12:01,170 --> 00:12:05,260 So these are the kind of things that we're trying to build up to and use a lot throughout the course 163 00:12:05,250 --> 00:12:05,580 . 164 00:12:05,580 --> 00:12:07,130 These sort of statements. 165 00:12:07,290 --> 00:12:12,640 Now this is how you can select rows or multiple rows based on some sort of condition. 166 00:12:12,630 --> 00:12:16,920 You can also use the subset function to do the exact same thing. 167 00:12:17,010 --> 00:12:24,480 So the subset function you say subset passing your data frame in this case it's actually empty cars 168 00:12:24,480 --> 00:12:25,120 . 169 00:12:25,470 --> 00:12:31,080 And then you basically do all these same commands except you don't have to worry about calling it off 170 00:12:31,080 --> 00:12:31,800 the data frame. 171 00:12:31,810 --> 00:12:36,270 Since you already passed the data frame into the subset function so it knows what you're actually talking 172 00:12:36,270 --> 00:12:36,800 about. 173 00:12:37,050 --> 00:12:46,440 So you can say something like MPG greater than 20 and empty cars. 174 00:12:46,470 --> 00:12:53,540 Sit ups use me and cylinders equals to six. 175 00:12:53,620 --> 00:12:58,170 So now return the same subset as this call. 176 00:12:58,200 --> 00:13:03,300 But notice what's nice about using subset if you prefer to use it that way is that you don't have to 177 00:13:03,390 --> 00:13:07,150 continually say empty cars dollar sign in the name of the column. 178 00:13:07,320 --> 00:13:12,030 Since you passed in the data frame it's subset function it already knows what you mean when you say 179 00:13:12,030 --> 00:13:15,110 something like MPG or seat y l. 180 00:13:15,120 --> 00:13:15,690 All right. 181 00:13:16,050 --> 00:13:22,410 So that's the basics of using a subset or this sort of bracket notation use whatever you feel most comfortable 182 00:13:22,410 --> 00:13:23,210 with. 183 00:13:23,220 --> 00:13:25,810 Let's go ahead and clear the console. 184 00:13:26,350 --> 00:13:29,480 We've already seen a few examples of how to select the multiple column names. 185 00:13:29,480 --> 00:13:32,710 Foolishest review them quickly. 186 00:13:32,860 --> 00:13:38,000 So I'll say have empty cars. 187 00:13:38,520 --> 00:13:43,100 Check the head of it one more time just so we can take a look at those column names we have. 188 00:13:43,110 --> 00:13:44,810 MPG cylinder. 189 00:13:44,800 --> 00:13:46,630 So I want to select multiple columns. 190 00:13:46,620 --> 00:13:48,300 There's a few ways I can do this. 191 00:13:48,630 --> 00:13:56,880 I can say empty cars brackets nothing comma and then a vector of either the numbers that relate to the 192 00:13:56,880 --> 00:13:57,980 columns I want. 193 00:13:58,140 --> 00:14:06,960 So for example if I want columns 1 2 and 3 in this case or B mpg c y l the ISP it returns back. 194 00:14:06,960 --> 00:14:13,860 If we scroll up here those street columns the other way of doing this we actually just saw is by saying 195 00:14:14,880 --> 00:14:18,820 passing in the actual names is a feel like MPG comma. 196 00:14:18,970 --> 00:14:21,370 See y while etc.. 197 00:14:21,810 --> 00:14:24,540 And if we scroll up to see this it just returns this column. 198 00:14:24,540 --> 00:14:25,780 So we asked for. 199 00:14:25,800 --> 00:14:26,540 So it's up to you. 200 00:14:26,550 --> 00:14:28,560 Usually I'll probably be using something like this. 201 00:14:28,650 --> 00:14:32,820 As you remember the order of the columns you just remember the name of the columns. 202 00:14:32,910 --> 00:14:37,520 Finally let's clear the consul and talk about dealing with missing data. 203 00:14:37,680 --> 00:14:41,740 So dealing with missing data is a pretty important skill to know especially when you're working with 204 00:14:41,740 --> 00:14:42,650 data frames. 205 00:14:42,780 --> 00:14:47,240 And there's a couple of useful built in functions to help you find missing data and check if there's 206 00:14:47,250 --> 00:14:49,330 missing data in your data front. 207 00:14:49,470 --> 00:14:56,280 So the way to do this is let's say we want to detect if there were any missing data points or an eight 208 00:14:56,280 --> 00:14:57,370 points. 209 00:14:57,370 --> 00:15:01,280 When I say any names we have no oil are missing data there. 210 00:15:01,360 --> 00:15:03,450 To wanted to take them anywhere in our data frame. 211 00:15:03,450 --> 00:15:04,670 How can we do that. 212 00:15:05,070 --> 00:15:08,420 We can say is not an A. 213 00:15:09,340 --> 00:15:16,320 And then passing your data frame So let's say we pass in empty cars if we pass this is an antique cars 214 00:15:16,320 --> 00:15:16,890 . 215 00:15:16,890 --> 00:15:20,610 We get back this data frame of boolean values. 216 00:15:20,820 --> 00:15:23,030 So we say false false fossils false. 217 00:15:23,250 --> 00:15:28,250 And notice we get all falses because there is no missing information in the state of frame. 218 00:15:28,380 --> 00:15:33,250 If for some reason one of these had an essay or no value we will get a true somewhere around the state 219 00:15:33,250 --> 00:15:34,300 of frame. 220 00:15:34,320 --> 00:15:40,350 So the way you can quickly check if you have any known or any are missing data in your data frame is 221 00:15:40,360 --> 00:15:41,930 bypassing that same argument. 222 00:15:41,940 --> 00:15:44,690 And we go in and clear the council. 223 00:15:44,940 --> 00:15:55,500 So we say is N.A. Asin your data frame and you can take advantage of the any function and that will 224 00:15:55,500 --> 00:15:58,250 check if any of those valleys is true. 225 00:15:58,710 --> 00:16:04,320 In this case none of those values are true so we get false back that if any single one of those values 226 00:16:04,340 --> 00:16:10,380 and that is that and a check on your data frame was true you would have gotten a true back as a report 227 00:16:10,380 --> 00:16:10,880 . 228 00:16:10,890 --> 00:16:17,130 So this is a nice little shortcut any isn't a data frame in order to check if you have any missing points 229 00:16:17,350 --> 00:16:19,610 anywhere in your data frame. 230 00:16:19,620 --> 00:16:23,660 Again you can expand this idea if you want to check if they're anywhere in a certain column. 231 00:16:23,730 --> 00:16:29,850 You can just pass in the set of your data frame pass in your column such as let's say for dealing with 232 00:16:29,940 --> 00:16:34,810 empty cars in pass in mpg. 233 00:16:36,430 --> 00:16:40,690 And if you wanted the whole empty car's data frame just say or any of them. 234 00:16:40,790 --> 00:16:41,290 No. 235 00:16:41,340 --> 00:16:47,210 Or N.A. if you want to replace missing data you can do that by taking advantage of this. 236 00:16:47,250 --> 00:16:50,640 Is that an a call so we can say something like this. 237 00:16:50,670 --> 00:17:00,260 Our data frame in this case will say D.S. will check is the whoops we'll put in a bracket and say is 238 00:17:01,540 --> 00:17:08,780 at a pass in your data frame and you can go ahead and pass in the values you want to replace. 239 00:17:08,800 --> 00:17:12,430 You can say replace all know values of physico. 240 00:17:12,550 --> 00:17:16,950 Usually you probably won't want to do such a broad command since it's going to do this for every single 241 00:17:16,950 --> 00:17:18,420 column in your data frame. 242 00:17:18,690 --> 00:17:19,720 But just keep that in mind. 243 00:17:19,720 --> 00:17:24,710 You can use the sort of is that in a notation check to replace null values. 244 00:17:24,900 --> 00:17:28,260 Do you only want it to do this for a single selected column. 245 00:17:28,260 --> 00:17:32,090 You'd use the exact same notation but you would pass and column names instead. 246 00:17:32,350 --> 00:17:35,790 So you would say something like for instance for the empty cars. 247 00:17:35,880 --> 00:17:41,740 Let's say we had a missing values in the MPG column you would say empty cars dollar sign mpg for the 248 00:17:41,730 --> 00:17:49,120 MPG column is dot and a you would say empty cars. 249 00:17:49,500 --> 00:17:52,780 Dollar Sign this case mpg. 250 00:17:52,980 --> 00:17:57,360 And then you would go ahead and replace that with whatever value you wanted to replace that with zero 251 00:17:57,370 --> 00:17:57,910 . 252 00:17:57,900 --> 00:18:08,220 What's nice is you can then also use something like mean of some column such as empty cars MPG to quickly 253 00:18:08,220 --> 00:18:12,860 replace no values with the average value of that column. 254 00:18:13,050 --> 00:18:16,200 And that's kind of a common method for dealing with missing data. 255 00:18:16,200 --> 00:18:20,490 Really depends on what your data looks like and what best practices are for that situation you're dealing 256 00:18:20,490 --> 00:18:21,060 with. 257 00:18:21,070 --> 00:18:26,490 But this sort of line can be really useful to quickly compute average or mean data into a column that's 258 00:18:26,490 --> 00:18:28,140 missing values. 259 00:18:28,140 --> 00:18:30,090 All right I hope that helped. 260 00:18:30,120 --> 00:18:35,370 Remember to use the notes as a reference in order to fully understand or reference a cheat sheet for 261 00:18:35,380 --> 00:18:36,450 this lecture. 262 00:18:36,880 --> 00:18:37,860 OK. 263 00:18:37,870 --> 00:18:39,780 Thanks everyone and I'll see you at the next lecture 27394