Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated: 1 00:00:00,660 --> 00:00:04,690 Hello everyone and welcome to the lecture on data frame basics. 2 00:00:04,800 --> 00:00:09,800 We've learned a lot about vectors and their two dimensional counterpart matrices. 3 00:00:09,810 --> 00:00:11,580 Now we're going to learn about data frames. 4 00:00:11,610 --> 00:00:17,730 One of the main tools for data analysis with our matrix inputs we're limited because all data inside 5 00:00:17,730 --> 00:00:23,700 of the matrix to be of the same data type with data frames we're going to be able to organize and mix 6 00:00:23,700 --> 00:00:27,090 data types to create powerful data structures. 7 00:00:27,330 --> 00:00:32,970 Let's go ahead and jump to our studio to begin exploring some built in data frames. 8 00:00:32,970 --> 00:00:33,350 All right. 9 00:00:33,390 --> 00:00:39,450 So here in our studio I'm going to go ahead and just push the right hand windows over to the right a 10 00:00:39,450 --> 00:00:40,440 little further. 11 00:00:40,440 --> 00:00:43,880 So we just see the console and the scripting screen. 12 00:00:44,580 --> 00:00:49,070 Let's start off by seeing and exploring some built in data frames. 13 00:00:49,470 --> 00:00:52,850 One of the first ones we're going to check comes from that state data set. 14 00:00:52,850 --> 00:00:57,410 So if you begin typing state into our studio you should see a bunch of pop ups. 15 00:00:57,420 --> 00:01:03,460 Let's go ahead and explore state x 7 7. 16 00:01:03,470 --> 00:01:06,840 And then when you click enter you'll notice a data frame pops up. 17 00:01:06,870 --> 00:01:08,720 This looks similar to a matrix. 18 00:01:08,760 --> 00:01:13,220 But notice how the rows and the columns are names. 19 00:01:13,530 --> 00:01:21,060 Now data frames can also have names rows and columns versus just having the number signifying what number 20 00:01:21,060 --> 00:01:23,080 column or what number row it is. 21 00:01:23,340 --> 00:01:28,230 This sort of labeling is going to be really useful or trying to access our data from this data frame 22 00:01:28,230 --> 00:01:28,770 . 23 00:01:28,770 --> 00:01:33,870 You can begin to think about data frames as sort of an X like an Excel spreadsheet. 24 00:01:33,930 --> 00:01:37,920 It's going to be really useful for working with large amounts of data. 25 00:01:37,920 --> 00:01:41,340 Let's go ahead and explore a few more builtin data frames. 26 00:01:41,380 --> 00:01:42,990 You know go ahead and clear the cons.. 27 00:01:43,260 --> 00:01:50,760 And I'm going to import U.S. personal expenditure so this data set consists of United States personal 28 00:01:50,760 --> 00:01:54,530 expenditures in the billions of dollars. 29 00:01:54,670 --> 00:01:57,290 It's a smaller data frame but you can see the same idea. 30 00:01:57,300 --> 00:02:00,730 We have a labeled row and some labeled columns. 31 00:02:00,870 --> 00:02:06,090 Now the label rows or the labeled columns to actually need to have string or character labels that can 32 00:02:06,090 --> 00:02:08,160 be numeric just like a normal index. 33 00:02:08,160 --> 00:02:15,150 So for example in other built in data frame is the women data frame you just begin to type women and 34 00:02:15,150 --> 00:02:21,210 this dataset gives the average height and weights for American women aged 30 to 39 of a small sample 35 00:02:21,210 --> 00:02:23,320 of them. 36 00:02:23,340 --> 00:02:28,230 So if we explore this data frame a little more we noticed we have two columns here height and weight 37 00:02:28,530 --> 00:02:33,760 with various values and each row is just a numeric indexed value. 38 00:02:33,760 --> 00:02:37,180 It doesn't actually have a label. 39 00:02:37,650 --> 00:02:42,270 You might be wondering well how do you get a list of all the available built in data frames that are 40 00:02:42,720 --> 00:02:43,600 in order to do that. 41 00:02:43,620 --> 00:02:51,480 You can just type in data close parentheses on that data function and you should get a new tab to pop 42 00:02:51,480 --> 00:02:52,280 up here. 43 00:02:52,470 --> 00:02:57,810 That will give you a list of all the data sets in the builtin data sets packaged with are. 44 00:02:57,810 --> 00:03:04,200 So if we scroll down here you'll see on the left the name that you need to import and a right description 45 00:03:04,230 --> 00:03:04,770 of it. 46 00:03:05,040 --> 00:03:11,970 So for instance let's say we just scroll down here we want to explore the world's telephone's will we 47 00:03:11,970 --> 00:03:14,730 just type in world phones to see that. 48 00:03:14,760 --> 00:03:23,130 So go ahead and type in world phones enter and so we start seeing some data about the number of phones 49 00:03:23,130 --> 00:03:27,280 in the world separated by columns of the different parts of the world. 50 00:03:27,600 --> 00:03:31,100 Something you've made noticed is that some of the data frames are really big. 51 00:03:31,350 --> 00:03:36,720 If you just want to take a peek at either the top or the bottom of the data frame you can use the head 52 00:03:36,750 --> 00:03:38,060 and tail functions. 53 00:03:38,250 --> 00:03:45,660 So for example remember we had that state DOT X 7:7 data frame which was quite large because it has 54 00:03:45,660 --> 00:03:47,490 all 50 states inside of it. 55 00:03:47,660 --> 00:03:57,930 We can go ahead and call ahead on State DOT x 7 7 and this will just return the first six rows of that 56 00:03:57,930 --> 00:04:06,070 data frame and we can also call tail state x 7 7 or just passing whatever data frame we want in they'll 57 00:04:06,080 --> 00:04:09,870 shoot the last six rows. 58 00:04:09,870 --> 00:04:15,900 Some other useful built in functions for working data frames are SDR and summary. 59 00:04:16,080 --> 00:04:22,320 We can use the TR function to get the structure of a data frame which gives information on the structure 60 00:04:22,320 --> 00:04:27,710 of the data frame and the data it contains such as variable names and data types. 61 00:04:27,720 --> 00:04:34,010 So for example if I call SDR which stands for structure and passing the data frame object. 62 00:04:34,260 --> 00:04:40,030 Let's go ahead and pass in that state X 7:7 click enter. 63 00:04:40,140 --> 00:04:42,950 Notice we begin to get some information about it. 64 00:04:43,160 --> 00:04:49,730 Like what kind of rows it has what kind of columns it has the dimensions etc.. 65 00:04:49,860 --> 00:04:55,430 There's another built in function which will give you some nice testicle data called summary. 66 00:04:55,560 --> 00:05:01,350 So summary is just a generic function that's used to produce results summaries of the results of various 67 00:05:01,410 --> 00:05:05,470 model fitting functions or just of objects like a data frame. 68 00:05:05,490 --> 00:05:10,460 So if you pass in your data frame here so summerise thought X 7:7 69 00:05:13,650 --> 00:05:17,540 notice we get a summary for each of the columns in our data frame. 70 00:05:17,550 --> 00:05:24,310 So for example the population summary gives you the minimum value the median value the mean values some 71 00:05:24,330 --> 00:05:27,290 court tile values as well as the maximum value. 72 00:05:27,780 --> 00:05:33,480 Later on in the course we're going to be using summary quite a bit for not just data frames but a whole 73 00:05:33,480 --> 00:05:35,790 host of functions and objects. 74 00:05:35,790 --> 00:05:40,040 So just keep that in mind since we've explored built in data frames. 75 00:05:40,050 --> 00:05:43,920 Let's go ahead and explore how we can create our own data frames now. 76 00:05:44,780 --> 00:05:48,770 OK we'll start off by creating a data frame from some vectors. 77 00:05:48,780 --> 00:05:51,550 Let's go ahead and make up some weather data. 78 00:05:51,570 --> 00:05:57,660 I'll start off by making a vector called days and we'll just have this be the days of the week which 79 00:05:57,660 --> 00:05:58,870 we've used before. 80 00:05:59,190 --> 00:06:03,300 So Monday Tuesday Wednesday 81 00:06:06,720 --> 00:06:16,710 Thursday and Friday we have Monday Tuesday Wednesday Thursday Friday for the days then we're going to 82 00:06:16,710 --> 00:06:24,200 make a temp sector and you can actually just quickly copy and paste this from the notes and I'm going 83 00:06:24,200 --> 00:06:26,250 to make up some temperatures. 84 00:06:26,280 --> 00:06:30,190 Let's say these are in Celsius now. 85 00:06:30,540 --> 00:06:31,240 OK. 86 00:06:31,620 --> 00:06:33,090 So we have my temperatures there. 87 00:06:33,130 --> 00:06:36,060 And let's make one final vector. 88 00:06:36,060 --> 00:06:39,390 This one's going to be a vector of logical values. 89 00:06:39,390 --> 00:06:44,430 So boolean values will indicate whether or not it rained that day. 90 00:06:44,460 --> 00:06:48,670 So we'll say it drained on Tuesday on Monday. 91 00:06:48,760 --> 00:06:54,770 Did it rain on Wednesday then in Reno Thursday and it rained again on Friday. 92 00:06:55,230 --> 00:07:05,400 Okay so we can go ahead and pass in these vectors into the data dot Frain function so we can go ahead 93 00:07:05,400 --> 00:07:16,020 and say days Tim rain and now we outputted a data frame and let's go ahead and actually save this as 94 00:07:16,020 --> 00:07:25,990 DMF variable data frame and then pass in the vector's days Tim brain. 95 00:07:26,700 --> 00:07:31,530 And then when we check out our data frame we'll notice that the column names have been automatically 96 00:07:31,530 --> 00:07:40,050 set to be the same as the name of the vector variable and by default the rows are just indexed numbers 97 00:07:40,050 --> 00:07:41,040 . 98 00:07:41,460 --> 00:07:49,430 Since we've now Cranor a data frame we can actually go ahead and call structure on it as TR RDF that 99 00:07:49,440 --> 00:07:52,920 will give us some information about the structure of the data in the data frame. 100 00:07:52,920 --> 00:07:56,780 Or we can even call a summary on that data frame. 101 00:07:57,490 --> 00:08:04,980 Okay that's it for the basics of what a data frame is and what it looks like and are the main things 102 00:08:04,980 --> 00:08:07,420 you should have been able to get out of this lecture. 103 00:08:07,620 --> 00:08:08,760 Are the following. 104 00:08:09,000 --> 00:08:15,120 You can type in data to get a output of all the data sets in the package data sets. 105 00:08:15,170 --> 00:08:16,300 You can play around with. 106 00:08:16,320 --> 00:08:18,720 Are these already built in for you. 107 00:08:18,720 --> 00:08:24,850 Also remember you can use SDR and summary to check out data frames whether you create them or they're 108 00:08:24,850 --> 00:08:25,750 built in. 109 00:08:26,010 --> 00:08:33,000 And finally you can create your own data frames for passing and vectors into the data dot frame function 110 00:08:33,000 --> 00:08:33,780 . 111 00:08:34,080 --> 00:08:34,950 OK. 112 00:08:35,010 --> 00:08:36,600 Again that's it for the basics. 113 00:08:36,600 --> 00:08:41,150 Up next we're going to learn about selection and indexing data frame elements. 114 00:08:41,160 --> 00:08:43,130 Thanks everyone and I'll see you at the next lecture 12140