Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated: 1 00:00:00,600 --> 00:00:06,140 Hello everyone and welcome to the our data frame's exercises solutions walk through lecture. 2 00:00:06,390 --> 00:00:11,700 This lecture is going to be programming through the solutions for the data frame's exercises and explaining 3 00:00:11,730 --> 00:00:12,820 as we go along. 4 00:00:13,110 --> 00:00:15,070 Let's jump to our studio and get started. 5 00:00:15,270 --> 00:00:15,570 OK. 6 00:00:15,570 --> 00:00:21,720 So here we are studio in the first exercise question was to recreate the following data frame by creating 7 00:00:21,720 --> 00:00:25,440 vectors and using that data frame function. 8 00:00:25,440 --> 00:00:30,570 So the data frame in question that we need to recreate is in the exercise notebook and have also printed 9 00:00:30,570 --> 00:00:31,710 it out here. 10 00:00:31,710 --> 00:00:34,410 So looks like we want three rows and three columns. 11 00:00:34,410 --> 00:00:38,890 Sam Frank and Amy are people as rows so this the index of the rows. 12 00:00:39,180 --> 00:00:43,180 And then we have an age column the weights column and a sex column. 13 00:00:43,220 --> 00:00:46,650 So as the instructions say we'll do it by creating vectors. 14 00:00:46,980 --> 00:00:53,380 So let's go ahead and start and go ahead and make a name vector to hold the names of the people. 15 00:00:53,820 --> 00:01:03,370 So that's going to be Sam Frank is our next one and Amy is our last one. 16 00:01:03,610 --> 00:01:07,200 You new single or double quotes here won't really make a difference. 17 00:01:07,200 --> 00:01:10,880 Next one will have is H. 18 00:01:11,610 --> 00:01:17,810 And I'll make that vector have 20 to 25 and 26. 19 00:01:18,210 --> 00:01:29,070 Then we have a wait's column so make a vector called a weight and that will carry 150 165 120. 20 00:01:29,070 --> 00:01:35,580 And then finally we have a six column for their gender and then we have them 21 00:01:38,390 --> 00:01:42,820 male again and then female. 22 00:01:42,840 --> 00:01:44,160 All right so we have a vectors. 23 00:01:44,160 --> 00:01:48,150 Now the question is how do we combine these into a data frame. 24 00:01:48,540 --> 00:01:55,350 Well we need to call our data that frame function and then I'm going to go ahead and say robot names 25 00:01:56,010 --> 00:02:03,570 is equal to the name vector and that's how we can assign those row names that index labeling to the 26 00:02:03,570 --> 00:02:04,980 name vector. 27 00:02:04,980 --> 00:02:10,020 Then I just need a pass in the columns that I want in this case are just the vectors. 28 00:02:10,020 --> 00:02:17,120 So I can say age weights and then that sex gender column. 29 00:02:17,220 --> 00:02:19,730 Let's go ahead and assign this to a data frame. 30 00:02:20,040 --> 00:02:26,760 If we can print out DSF And if we just read this we get the exact same result looks. 31 00:02:26,820 --> 00:02:33,060 If you've got a match fantastic may go ahead and clear the con.. 32 00:02:33,060 --> 00:02:38,610 One other thing I want to mention before we continue onto the next exercise is in case you couldn't 33 00:02:38,610 --> 00:02:44,460 figure out how to actually set the names of the row row that names using this input data frame function 34 00:02:44,790 --> 00:02:51,840 you could have also set it using the row names function that we've seen before row names and then what 35 00:02:51,840 --> 00:02:56,100 we do is just passing your data frame and then assign a vector of names. 36 00:02:56,100 --> 00:03:04,500 So for example if we had different names such as ABC just like we did for matrices you can use this 37 00:03:04,620 --> 00:03:11,230 same functionality knows another way you could have set those names for the rows. 38 00:03:11,250 --> 00:03:11,830 All right. 39 00:03:12,060 --> 00:03:15,280 So we have age weight sex and ABC. 40 00:03:15,300 --> 00:03:18,030 Let's go ahead and continue on. 41 00:03:18,070 --> 00:03:26,390 Going to go ahead and clear this text here and it clear the council and put in the next exercise question 42 00:03:26,390 --> 00:03:26,980 . 43 00:03:27,060 --> 00:03:30,620 So the next exercise question was the check of empty cars is a data frame using. 44 00:03:30,630 --> 00:03:32,300 Is it a frame. 45 00:03:32,310 --> 00:03:38,080 So again empty cars is a built in data frame in our So it's go and just check the head of it. 46 00:03:38,100 --> 00:03:42,840 You don't need to import any libraries or do anything you see a type empty cars it will automatically 47 00:03:42,840 --> 00:03:45,350 know that you're referencing that data frame. 48 00:03:45,360 --> 00:03:49,620 Quick reminder do you want to see what other data is available for you that's built in. 49 00:03:49,620 --> 00:03:56,520 You can say data as a function and this little pop up will show up with names of builtin matrices data 50 00:03:56,520 --> 00:03:58,850 friends vectors etc.. 51 00:03:58,850 --> 00:04:00,960 We're going to close that now. 52 00:04:01,410 --> 00:04:07,830 So we want to check if MT Kerr's is a data frame so we can always check if an object is a particular 53 00:04:07,830 --> 00:04:10,890 type of class or a particular type of data structure etc.. 54 00:04:11,020 --> 00:04:13,540 But using is datt methodology. 55 00:04:13,560 --> 00:04:18,870 So that is that we can then we just say whatever we're actually checking for in this case we're checking 56 00:04:18,870 --> 00:04:20,470 for is that data frame. 57 00:04:20,670 --> 00:04:26,000 And we just pass in empty cars and it returns True which is good because IndyCar has a built in data 58 00:04:26,000 --> 00:04:26,550 . 59 00:04:26,820 --> 00:04:33,130 Next exercise was to use as that data frame to convert a matrix into a data frame. 60 00:04:33,330 --> 00:04:41,100 So just like we have these is options we also have as options and as options will basically try to convert 61 00:04:41,520 --> 00:04:45,200 from one object or a data type to another. 62 00:04:45,300 --> 00:04:47,970 So we're going to say as well actually. 63 00:04:47,970 --> 00:04:58,710 First off we want to actually set up our matrix so the matrix in this case is this M-80 we sign that 64 00:04:58,710 --> 00:05:06,730 matrix and then we can say as that data frame pass in M-80 and we get a data frame back. 65 00:05:06,730 --> 00:05:12,360 So if I just say amitie or Matt by itself notice a difference in the output display here we can see 66 00:05:12,360 --> 00:05:18,270 it's a matrix due to the bracket notation in the kitting rows and columns just by index or array numbers 67 00:05:18,270 --> 00:05:18,610 . 68 00:05:18,660 --> 00:05:25,880 Here we can see the data frame actually has built in column names and a builtin row naming scheme. 69 00:05:25,890 --> 00:05:26,210 All right. 70 00:05:26,250 --> 00:05:28,530 Moving on to the next exercise question. 71 00:05:28,860 --> 00:05:34,790 It was to set the builtin data for an empty cars as a variable DFAC if I clear the council. 72 00:05:34,980 --> 00:05:40,380 All we had to do for this step was quite simple just say the F is empty cars and we're going to be doing 73 00:05:40,380 --> 00:05:43,880 is referring to ADF for the rest of the questions. 74 00:05:44,010 --> 00:05:49,240 So the next one was to display the first six rows of DLF. 75 00:05:49,260 --> 00:05:51,440 Question number five how do we actually do that. 76 00:05:51,450 --> 00:05:56,510 Well we can just call ahead on the path though automatically display the first six rows. 77 00:05:56,510 --> 00:06:00,430 Do you want to display a certain number of rows from the top of your data frame. 78 00:06:00,510 --> 00:06:06,060 You can specify a second argument and head which is just an integer saying OK only display the first 79 00:06:06,060 --> 00:06:09,420 two rows six seven rows etc.. 80 00:06:09,660 --> 00:06:17,040 Next question we had to answer was What does the average MP G or mpg value for all the cars. 81 00:06:17,460 --> 00:06:22,000 Well north to answer this question let's go in and check. 82 00:06:22,030 --> 00:06:27,300 Looks like we have an MPG column number the way one way of calling columns off a data frame is just 83 00:06:27,300 --> 00:06:29,220 by using the dollar sign. 84 00:06:29,220 --> 00:06:34,110 So using that methodology I can get a vector of the values and have a vector of values. 85 00:06:34,110 --> 00:06:41,370 It means I can just call me in and say the F mpg. 86 00:06:42,030 --> 00:06:47,550 And there we have it's around twenty point one miles per gallon is the average MPG value for all the 87 00:06:47,550 --> 00:06:48,800 cars. 88 00:06:48,870 --> 00:06:50,800 Go in and go onto the next question. 89 00:06:51,600 --> 00:06:53,960 OK so exercise seven. 90 00:06:53,970 --> 00:06:58,040 Question number seven was to select the rows for all cars have 6 cylinders. 91 00:06:58,090 --> 00:06:59,850 There's a couple of ways we can do this. 92 00:06:59,880 --> 00:07:06,570 One way is through bracket notation where we can just say since we have a data frame as empty cars DSF 93 00:07:07,110 --> 00:07:13,060 specify the cylinder column equals six. 94 00:07:13,110 --> 00:07:14,770 And then we have to add an extra comma. 95 00:07:14,850 --> 00:07:17,970 Since we're looking for all the rows where that's true. 96 00:07:17,970 --> 00:07:23,740 And then that will return where the data frame cylinder column has equality with six. 97 00:07:23,760 --> 00:07:29,520 So that's why these bracket notation we can also use the subset function to do this. 98 00:07:29,640 --> 00:07:38,850 So we can say subsets pass in our data frame and then say cylinder equals to 6 and they'll produce the 99 00:07:38,850 --> 00:07:40,090 exact same result. 100 00:07:40,320 --> 00:07:41,740 Either method is correct. 101 00:07:41,760 --> 00:07:46,890 This is the method shown in the solution notebook but you could have also done subset the cylinders 102 00:07:47,090 --> 00:07:49,450 six is a couple of other ways to do this. 103 00:07:49,470 --> 00:07:55,350 And later on later in the course that is all there and how to use the player library to also filter 104 00:07:55,350 --> 00:07:57,800 out results using some special functions. 105 00:07:57,940 --> 00:08:01,020 For now either of these two methods would have been correct. 106 00:08:01,020 --> 00:08:09,680 Moving on to the next exercise we had to select the columns am gear and carb from the data frame. 107 00:08:09,690 --> 00:08:11,510 So how do we actually do that. 108 00:08:11,690 --> 00:08:12,750 We clear the console. 109 00:08:12,780 --> 00:08:19,650 We know if we want to select just one column we can say bracket notation comma and then the name of 110 00:08:19,650 --> 00:08:26,640 the column such as am and I'll return those vector values if we want several columns so we can just 111 00:08:26,640 --> 00:08:29,810 pass any vector of the column names. 112 00:08:29,910 --> 00:08:39,780 So we want AM year car and there we have it we scroll up to actually see this. 113 00:08:39,840 --> 00:08:42,260 This is the resulting data frame. 114 00:08:42,330 --> 00:08:46,710 So you get those three columns back along with their real names that are associated with each of those 115 00:08:46,710 --> 00:08:48,300 values. 116 00:08:48,300 --> 00:08:52,590 There's a couple of other ways to do this but this is probably the most straightforward as far as you 117 00:08:52,590 --> 00:08:55,520 see him bracket notation indexing to do this. 118 00:08:55,530 --> 00:09:02,280 Moving on to the next question that was to create a new column called performance which is calculated 119 00:09:02,280 --> 00:09:04,500 by horsepower divided by weight. 120 00:09:04,740 --> 00:09:06,940 Let's go ahead and do that. 121 00:09:07,320 --> 00:09:08,930 Can it clear the council. 122 00:09:09,720 --> 00:09:13,150 So how do we actually create a new column with a data frame. 123 00:09:13,470 --> 00:09:14,740 Well there are several ways to do it. 124 00:09:14,760 --> 00:09:21,180 The easiest is just by specifying that column as if it already exists and then assigning it some values 125 00:09:21,180 --> 00:09:21,270 . 126 00:09:21,270 --> 00:09:24,240 In this case we want to assign horsepower the value by weight. 127 00:09:24,300 --> 00:09:30,870 So just go ahead and call those columns horse power after they reframe divided by weight. 128 00:09:31,340 --> 00:09:33,790 And let's go ahead and check the head of our data free now. 129 00:09:34,350 --> 00:09:40,830 And notice we have the new performance column and this will lead us into our next question and our next 130 00:09:40,830 --> 00:09:46,140 question notice that the performance column has several decimal place precision so it looks it goes 131 00:09:46,140 --> 00:09:47,940 up to five decimal places. 132 00:09:48,090 --> 00:09:53,820 We want to figure out how to use round to reduce this accuracy to only two decimal places. 133 00:09:53,820 --> 00:09:55,360 And it says check help round. 134 00:09:55,440 --> 00:09:56,580 Let's go ahead and do that. 135 00:09:56,730 --> 00:10:04,560 So if we haven't seen round before we can say help round her and we get this nice help documentation 136 00:10:04,650 --> 00:10:06,890 on the rounding of numbers. 137 00:10:06,990 --> 00:10:12,540 There's several functions to help us round numbers but we're looking just for round which round is the 138 00:10:12,540 --> 00:10:16,170 value in the first argument to the specified number of decimal places. 139 00:10:16,170 --> 00:10:24,420 So if we go ahead and copy and paste the documentation line looks like this and we end up having is 140 00:10:24,780 --> 00:10:26,250 two arguments here. 141 00:10:26,250 --> 00:10:27,600 X and digits. 142 00:10:27,610 --> 00:10:35,280 So X is the numeric vector and what digits represents is the number of decimal places that we want to 143 00:10:35,280 --> 00:10:36,980 use. 144 00:10:36,990 --> 00:10:42,180 Let's go ahead and shift this over to the right now that we know how to use round. 145 00:10:42,210 --> 00:10:46,890 Let's go ahead and reassign performance. 146 00:10:46,890 --> 00:10:51,170 So say performance is going to be equal to. 147 00:10:51,210 --> 00:10:55,600 And we can go ahead and say DMF performance again. 148 00:10:55,770 --> 00:11:01,530 And in this case what we're going to do is use round to pass in round. 149 00:11:01,530 --> 00:11:06,270 And the second argument we pass on is to which is the digits argument to make that really clear we can 150 00:11:06,270 --> 00:11:09,600 just say digits equals 2. 151 00:11:10,320 --> 00:11:15,240 And now if I check the head of my data frame I notice that my performance column has been truncated 152 00:11:15,330 --> 00:11:17,640 or rounded off to two digits. 153 00:11:17,640 --> 00:11:20,250 So it's not a straight truncation it's just a rounding off. 154 00:11:20,250 --> 00:11:26,280 So for example thirty point three four six gets rounded to thirty point three five. 155 00:11:26,280 --> 00:11:29,580 Let's go ahead and move on to the next question. 156 00:11:29,610 --> 00:11:31,540 Next question Frigo and informant. 157 00:11:31,560 --> 00:11:40,080 This was what is the average MPG for cars that have more than 100 horsepower and a weight value of more 158 00:11:40,080 --> 00:11:41,500 than 2.5. 159 00:11:41,850 --> 00:11:43,630 Let's go ahead and figure this out. 160 00:11:44,130 --> 00:11:47,330 There's a couple of ways you can solve this. 161 00:11:47,400 --> 00:11:49,880 Also the first method using subset. 162 00:11:50,220 --> 00:11:56,580 So our first challenge is to grab the subset of the data frame where we have more than 100 horsepower 163 00:11:56,610 --> 00:11:59,310 and a weight value of more than two point five. 164 00:11:59,310 --> 00:12:04,960 So I can say subset pass in my data frame and then pass in my condition. 165 00:12:04,960 --> 00:12:08,390 So in this case want a horse power greater than 100. 166 00:12:08,980 --> 00:12:17,050 And so using that logical operator I want weight to be also greater than 2.5. 167 00:12:17,730 --> 00:12:24,480 So if I go ahead and call that subset I get back to a subset of the data frame where this is true and 168 00:12:24,480 --> 00:12:29,290 I can actually call columns off of that subset command. 169 00:12:29,310 --> 00:12:35,770 So when you go in and clear this from that subset command I can call a column off of it. 170 00:12:35,880 --> 00:12:45,600 MPG which means I can take that whole statement and pass it into the mean function. 171 00:12:46,050 --> 00:12:51,210 And there you have sixteen point eight six etc. which is the average miles per gallon for cars that 172 00:12:51,210 --> 00:12:56,430 have more than 100 horsepower and a weight value of 2.5. 173 00:12:56,430 --> 00:13:00,700 That's how you can solve this question using the subset function. 174 00:13:00,720 --> 00:13:04,100 Now we could also bracket notation to do this. 175 00:13:04,190 --> 00:13:06,930 I'll go ahead and show you how we could have done that. 176 00:13:07,170 --> 00:13:13,750 We can say DSF and in brackets pass what the actual conditions you want. 177 00:13:14,190 --> 00:13:21,480 So this gets a little messier because we have to specify DFI dollar signs but it's essentially the same 178 00:13:21,480 --> 00:13:21,940 logic. 179 00:13:21,990 --> 00:13:32,650 We're saying T.F. horsepower greater than 100 and DPF weights greater than 2.5. 180 00:13:33,150 --> 00:13:40,930 And then what we can do off of this is put a comma call mpg Whoops. 181 00:13:41,070 --> 00:13:45,550 And then we just see that we get the exact same results. 182 00:13:45,750 --> 00:13:54,190 So I can call I mean on this entire thing and this is how you would do it using bracket notation. 183 00:13:54,720 --> 00:13:59,320 Personally subset looks a lot cleaner and has a lot more readable to me personally. 184 00:13:59,460 --> 00:14:03,130 But if you really like bracket notation you could have also done it this way. 185 00:14:03,600 --> 00:14:08,250 As I mentioned earlier later on we'll learn how to use that the player library to try to clean up these 186 00:14:08,250 --> 00:14:12,180 sort of filter instructions with a nice clean syntax. 187 00:14:12,180 --> 00:14:18,320 Finally the last question was this what is the mpg of the Hornet sport about. 188 00:14:18,390 --> 00:14:20,220 So how do we actually find that. 189 00:14:20,580 --> 00:14:28,580 Well I can't get my data frame and then just pass in the name of that car on it. 190 00:14:28,590 --> 00:14:31,660 Spore about karma. 191 00:14:31,800 --> 00:14:37,920 So I pass this first because that's the actual name comma because I want all the cars for that. 192 00:14:37,950 --> 00:14:41,590 So if I just do this they'll return the horn at sport about Roe. 193 00:14:42,030 --> 00:14:47,970 And if I want the mpg of that I can just say dollar sign mpg and there we have eighteen point seven 194 00:14:48,040 --> 00:14:51,140 The MPG of the Hornet sport about car. 195 00:14:51,180 --> 00:14:51,860 OK. 196 00:14:52,110 --> 00:14:56,870 That's it for this lecture on the solutions walk through for the data frames exercise. 197 00:14:56,880 --> 00:15:02,130 If any of that was unclear makes you reference the notebook work through the exercises again or reference 198 00:15:02,130 --> 00:15:05,600 the data frames lectures from the data frame section of the course. 199 00:15:05,610 --> 00:15:07,390 Thanks everyone and I'll see you at the next lecture 21166