WEBVTT 00:00.480 --> 00:01.410 Hi everybody. 00:01.770 --> 00:07.410 So in this video we learn how to select rows with suspect brackets in the same way as the last. 00:07.410 --> 00:10.760 The lesson we learned how to select the columns as Bap. 00:10.770 --> 00:14.130 However by selecting arrows is back brackets them. 00:14.130 --> 00:15.680 This is actually not advisable. 00:15.710 --> 00:22.200 And I show you in this lesson and the next lessons that there are better alternatives to do this. 00:22.200 --> 00:26.130 How about pen does this the complex topic and far for many tasks. 00:26.190 --> 00:32.580 There are many ways to do it and in most cases and there are good ways to do something and the very 00:32.580 --> 00:36.730 best ways to do something that are also dangerous and can lead to mistakes. 00:36.730 --> 00:43.620 So with large data sets this hundred thousands of millions of rows you even do not get a read that you 00:43.620 --> 00:47.940 have done something wrong or something which was actually not your intention. 00:48.060 --> 00:53.700 And I think it's also important to show you the pitfalls and make you about potential problems in order 00:53.700 --> 00:57.480 to prevent you from making mistakes that are actually not necessary. 00:58.410 --> 01:05.640 So first of all as always let's import peanuts and we import our Titanic data set and we check in the 01:05.640 --> 01:11.980 first rows and now we assume we want to select them only the very first peasants are here with them. 01:11.990 --> 01:24.310 The row labeled 0 and in the same way as with columns we just try to select our first row here by passing 01:24.340 --> 01:31.290 our row label 0 within this brackets and let's see what we get here and we get an error message. 01:31.300 --> 01:40.620 Key arrow 0 so actually paste and does not know if you want to select a row or column and it searches 01:40.710 --> 01:42.370 for the column labeled 0. 01:42.390 --> 01:49.140 And that is actually no column with the column labeled 0 and consequently passing the drops us here 01:49.140 --> 01:57.010 an error message but what we can do here actually we can make a trick and slice from 0 to 1 and let's 01:57.010 --> 01:57.930 see what we get here. 01:57.940 --> 02:02.550 So we get here the first peasant service them index labeled 0. 02:02.800 --> 02:08.830 And yeah actually this works in the same mass slicing a list or an umpire Ray now so so-called position 02:08.830 --> 02:09.730 based slicing. 02:09.740 --> 02:16.060 So the starting point is including and the ending point is excluding so we are slicing our data frame 02:16.510 --> 02:17.840 from the index label. 02:17.850 --> 02:28.830 0 intuiting onto the index label 1 excluding and this way we can make different slices here so we can 02:28.830 --> 02:29.660 also slice. 02:29.670 --> 02:33.180 Let's say from 3 to 5. 02:33.600 --> 02:39.990 So we would expect to have their overs index label 3 and 4 and excluding 5. 02:40.050 --> 02:40.730 That's correct. 02:41.930 --> 02:48.530 And we can also slice for the very first 10 rows here 02:52.320 --> 02:57.980 so the very first 10 rows excluding the rovers index label 10. 02:58.080 --> 03:05.610 So we get here from 0 to 9 and this is actually the same as having here the head method 03:12.150 --> 03:13.080 and applying 10. 03:13.080 --> 03:15.480 So give me the first 10 rows. 03:15.630 --> 03:22.630 So from 0 to 9 and of course we can also slice 03:25.930 --> 03:39.210 the very last 10 rows so it can slice from minus 10 to the very last row so you can see in the last 03:39.210 --> 03:47.010 10 rows and as before applying the tail method with an equals 10 would give us and the same results 03:48.600 --> 03:53.630 so this was a quick overview of our slicing rows with the spat brackets. 03:53.640 --> 03:57.110 And as I said before this is not a good way to do it. 03:57.120 --> 04:00.410 So we will see in the next lessons how to do it in the correct way. 04:00.410 --> 04:01.770 So I hope to see you there by.