WEBVTT 00:01.390 --> 00:05.520 In this video we learn how to change our ROE index in a data frame. 00:05.560 --> 00:07.930 And we have already seen that by default. 00:07.930 --> 00:13.540 We have a range index starting from zero but in many cases it might be beneficial to ask a different 00:13.540 --> 00:18.050 DRO index that enables us to label based indexing. 00:18.210 --> 00:25.600 So but first of all let's import pandas and we import our Summer Olympic Games data set and start in 00:25.600 --> 00:27.000 the variable summer. 00:27.820 --> 00:34.060 And we have a look at the first five rows and we have a look at the index. 00:34.070 --> 00:41.310 So we have a range index starting from zero and we have already learned that we can change the index 00:41.340 --> 00:48.810 and select a column as the index and imparting our data with the read CSP method so we can say here 00:49.590 --> 00:56.870 index column equals and let's say we want to have the athlete column as our index column so we can pass 00:56.880 --> 00:57.480 athlete 01:02.970 --> 01:09.570 and now we have few on the left side our athlete column s and our index off often the data frame and 01:09.570 --> 01:11.280 we can check the index here 01:14.200 --> 01:19.580 and here we can see our index is now at the athletes column. 01:19.600 --> 01:25.360 So now what if we are not happy with our index and we want to change it back to the Range Index and 01:25.450 --> 01:29.450 therefore we have the method reset index. 01:29.650 --> 01:39.760 So let's just try it out here and we can see here on the left side we again half our Range Index and 01:39.790 --> 01:41.270 our athlete column. 01:41.310 --> 01:49.060 Tom Speck as a column so it's not any more index but a normal column of our data frame and we can also 01:49.060 --> 01:57.030 check here the parameters and we have one parameter drop equals force. 01:57.310 --> 02:01.020 So this is set by default to force. 02:01.180 --> 02:03.090 So let's try it out again. 02:03.190 --> 02:11.200 And what if he set this to true here then we completely drop our index that we had before. 02:11.200 --> 02:17.070 So before we head to the athlete column as our index and if he said that the parameter dropped too true 02:17.110 --> 02:23.380 we completely drop our index and replace it by the range and next and you can see here there is no athlete 02:23.380 --> 02:28.810 column in our data frame so that's uh in this case actually not what we uh want. 02:28.840 --> 02:33.470 So we change it back to false. 02:33.480 --> 02:39.250 Now again we have our athlete column and our data frame and also in the Reset index method and there 02:39.260 --> 02:45.360 is and the in-place parameter so by default it's set to fall. 02:45.390 --> 02:51.510 So we do not change our original data frame and if we want to change our original data frame and change 02:51.510 --> 02:57.560 the index back to the Range Index then we have to set of the in-place parameter to true. 02:58.650 --> 03:04.650 So now if you run here so now we can see here that we do not get a view on a copy. 03:04.640 --> 03:10.470 So this is a good indication that we changed our original data frame and let's check with the method 03:11.050 --> 03:12.860 and then we can see here on the left side. 03:12.870 --> 03:20.130 Again we have a Range Index and still we have our athlete column as a normal column of our data frame. 03:20.190 --> 03:25.370 So now let's assume that before we we are not happy with the athlete column as in the index. 03:25.410 --> 03:31.550 But now we want to have the year column as our index and there are actually two ways to do it. 03:31.550 --> 03:39.140 So either we do this already when we import the data and set here the index card to year or if we have 03:39.140 --> 03:40.050 already. 03:40.510 --> 03:47.810 Well if we have already imported our data that is in the set index method to do this and let's press 03:47.810 --> 03:50.360 shift tap here. 03:50.360 --> 03:56.450 So the set next method to set to the data frame index using one or more existing columns. 03:56.450 --> 04:01.700 So what we can do here actually we can pass them the column that we like to have as the index. 04:01.700 --> 04:09.790 So in this case here in quotation marks and let us see what we get here then you're on the left side 04:09.820 --> 04:13.750 as in the index we have in the year column. 04:13.860 --> 04:20.550 All right so let's again press shift tap here and that's a parameter called drop. 04:20.670 --> 04:24.500 And by default drop is set to true. 04:24.540 --> 04:30.300 That means if we accept that the year column mass index it will be removed as an ordinary column. 04:30.510 --> 04:37.530 And if we change this and accept the drop curve parameter to force the year column and not be removed 04:37.590 --> 04:38.470 as a column. 04:38.490 --> 04:43.130 So in the end we have the year column twice as an index and as an ordinary column. 04:43.290 --> 04:48.170 So let's check this out. 04:48.290 --> 04:50.090 So by default it's set to true 04:53.020 --> 05:02.730 and we can see that we do not have a year column as column but if we accept this to false then we still 05:02.730 --> 05:06.750 have the year column as column and s index. 05:06.840 --> 05:09.060 So let's change it back here to true 05:11.750 --> 05:16.890 and uh let's set up the in-place parameter to a true 05:21.610 --> 05:30.330 and let's say again our data frame and now we have from the year column as index and we actually removed 05:30.410 --> 05:32.040 the column from the column. 05:32.050 --> 05:38.650 So there's no column year and now we can also check if the index is unique. 05:38.870 --> 05:44.960 So I would assume that we get a false because already here we can see that we have a duplicate value 05:44.990 --> 05:49.250 so eighteen hundred ninety six. 05:49.340 --> 05:49.880 All right 05:52.820 --> 05:59.660 so now we have learned how to change the complete index so we can reset our index to a range index and 05:59.660 --> 06:02.650 we can set a complete column as our new index. 06:02.680 --> 06:10.300 But what if you want to change only one element of our index and we can try this so let's have a look 06:10.300 --> 06:13.000 here at our summer index. 06:13.000 --> 06:16.560 So we fear our year columns index and the few select them. 06:16.600 --> 06:22.800 The first element so here it's in the first element 896. 06:22.840 --> 06:30.430 And if you try to set a new values of for example eighteen hundred ninety four when we get here an error 06:30.430 --> 06:39.870 message index does not support mutable operations so with other words the index objects immutable and 06:39.870 --> 06:45.990 we can also tried to set a new value for all elements of the index object so-called broadcasting so 06:45.990 --> 06:50.410 we can say that for all elements in our index we want to set them the new value. 06:50.550 --> 06:58.060 So the string here before 2016 and we can try this and again we get here an error message. 06:58.690 --> 07:05.880 So this doesn't work but what we can do we can pass a complete list or sequence and that has exactly 07:05.880 --> 07:08.960 the same amount of elements as our index. 07:08.960 --> 07:10.340 Then as the new index. 07:10.340 --> 07:13.710 So first of all let's check the size of our index. 07:13.710 --> 07:18.810 So we have thirty one thousand one hundred sixty five elements in our index. 07:18.810 --> 07:26.600 So if we pass a list or a sequence with is thirty one thousand one hundred sixty five new elements as 07:26.610 --> 07:29.250 our new index then this should work. 07:29.250 --> 07:35.070 And therefore we're creating here a list with thirty one thousand one hundred sixty five elements and 07:35.070 --> 07:36.930 we want to have a new label. 07:36.930 --> 07:41.280 And starting from medal number one better number two medal number three and so on. 07:41.280 --> 07:47.370 So we want to Numa rate our medals from medal number one and two medal number thirty one thousand one 07:47.370 --> 07:48.390 hundred sixty five. 07:48.390 --> 07:53.360 And therefore we use here a list comprehension. 07:53.430 --> 08:01.190 So for Ian Raines one to thirty one thousand one hundred sixty five including we name our elements and 08:01.230 --> 08:04.620 our new list medal number one and number two and so on. 08:04.620 --> 08:07.860 So now let's create our list here and let's have a look 08:10.620 --> 08:12.620 so here you can see our list. 08:12.630 --> 08:21.440 So it starts with medal number one and yeah in the end we're ending with a medal number thirty one thousand 08:21.440 --> 08:23.190 one hundred sixty five. 08:23.260 --> 08:30.590 And what we now can do we can say OK we want to set our index to a new index so new index here is our 08:30.590 --> 08:34.910 list and let's try out if this works. 08:34.910 --> 08:35.350 All right. 08:35.360 --> 08:42.820 We do not get an error message and let's check in the first five rows and the last five rows and two 08:42.850 --> 08:49.690 we can see that we have set our list new index as the index of our data frame and also check the last 08:49.690 --> 08:50.440 five for us. 08:51.070 --> 08:56.210 So starting from medal number one to the medal number thirty one thousand one hundred sixty five. 08:56.350 --> 09:00.670 And we can also check if our row index only has a unique value. 09:00.670 --> 09:03.430 So I would assume so yes. 09:03.450 --> 09:03.860 All right. 09:03.870 --> 09:12.420 So and that's a in addition a attribute called index stock name and let's check if we have a name already 09:12.420 --> 09:16.240 for our index and here we run this. 09:16.280 --> 09:22.460 And we see nothing so that's them at this point no name for our new index button but we can set a new 09:22.460 --> 09:26.810 name by assigning for example a medal number. 09:26.810 --> 09:36.980 So let's try this out and let's again check in the first five rows here and now here we can see that 09:37.130 --> 09:39.140 we have a name for our index. 09:39.170 --> 09:44.640 So we have medal number and now if we apply again the reset index method 09:49.150 --> 09:54.970 we can see here on the left that we again have a range index and our index that we had before here medal 09:54.970 --> 09:55.440 numbers. 09:55.450 --> 10:02.140 Now a column of our data frame with the column label medal number. 10:02.190 --> 10:06.600 All right so now we are finished service this video and I hope to see you on the next video by.