WEBVTT 00:00.810 --> 00:07.470 In this video we learn how to remove complete columns from a data frame and falsify the import pandas 00:08.160 --> 00:10.590 and import our Summer Olympic Games. 00:10.650 --> 00:11.610 Metadata friend 00:14.460 --> 00:17.210 and have a look at the first five rows. 00:17.760 --> 00:24.350 And here we can see that we have the columns that sparked discipline and we went and let us assume we 00:24.350 --> 00:26.390 only want to keep the column event. 00:26.420 --> 00:32.260 Because then the information in the spot column and disciplined column are a bit redundant actually 00:34.660 --> 00:42.580 so first of all we tried to remove the spot column and therefore we can use them the drop method and 00:42.580 --> 00:43.540 then the drop method. 00:43.540 --> 00:50.440 We can pass the label off the column we want to remove to the parameter columns and let's see what we 00:50.440 --> 00:53.750 get here. 00:53.830 --> 00:56.830 We get here a data frame without them these sports column 01:05.810 --> 01:11.490 All right so let's check again our data frame and here we can see that the sponsor column is still in 01:11.490 --> 01:18.930 our data frame so also the drop method has the in-place parameter of it's us by default set to false 01:19.440 --> 01:24.930 and if you want to change our data frame we have to set them the in-place parameter to true. 01:25.200 --> 01:31.200 But before we said we want to drop them the sports column and the discipline column and we can of course 01:31.230 --> 01:39.500 also remove multiple columns within the drop method and we can do this by passing a list of column labels 01:39.500 --> 01:42.000 to the parameter columns. 01:42.440 --> 01:43.460 So let's check here 01:46.740 --> 01:51.220 so here we can see that we remove the column sport in the column discipline. 01:51.220 --> 01:56.660 And now if you want to make this in place and change our rates or data frame we have to set that the 01:56.660 --> 02:02.030 in-place parameter to true and let's check again our data frame 02:05.190 --> 02:12.590 and now we can see that now in our data frame there is no sport and no discipline column and let's assume 02:12.590 --> 02:17.740 we want to drop also the event column then we again use the drop method. 02:17.960 --> 02:25.580 But instead of passing the event column to the parameter column we can also pass event to the parameter 02:25.580 --> 02:27.380 label so let's check here. 02:27.410 --> 02:33.080 So before we passed along our columns that we wanted to drop to the perimeter columns but we can also 02:33.080 --> 02:39.560 pass them this to labels and define the access from which we want to drop our label. 02:39.560 --> 02:49.700 And in this case it's either we can use 1 4 columns so let's secure now where we have removed the event 02:49.750 --> 02:57.910 or we can use in quotation marks columns and is missing here 03:01.590 --> 03:03.420 so that's actually the second alternative. 03:03.420 --> 03:10.200 So either we pass our column labels to the pyramid columns we posit to the perimeter labels and define 03:10.260 --> 03:17.050 the axis as columns or as one and again we have to separate the in-place parameter to true 03:22.210 --> 03:26.420 and second again our data frame. 03:26.430 --> 03:33.480 So now we have removed uh the spots column the discipline column and the event column and that's another 03:33.720 --> 03:35.260 way to remove a column. 03:35.310 --> 03:40.910 This the keyboard delete and then selecting the event column from our summer data frame. 03:40.920 --> 03:42.800 But I wouldn't recommend this one. 03:42.800 --> 03:48.740 So I would definitely go for the top method here and use the drop method to drop a column. 03:49.800 --> 03:54.350 So now let's import our summers the data again. 03:54.510 --> 03:57.080 So this is the original one. 03:57.360 --> 04:04.350 And before we drop them the columns sport discipline and event but instead of dropping the two columns 04:04.380 --> 04:12.750 we could also explicitly select the other columns so we could select a year city athlete country gender 04:12.810 --> 04:16.170 and medal and we can do this service to the local operator. 04:17.120 --> 04:23.180 So we select all rows and then we pass a list of column labels that we want to keep in our summer data 04:23.180 --> 04:23.600 frame. 04:23.600 --> 04:32.200 So yes city athlete country gender and medal and we override our summers data frame so our new data 04:32.200 --> 04:39.460 frame summer data frame contains the column C a city athlete country tandem medal and it does not contain 04:39.490 --> 04:44.260 the columns sport discipline and event. 04:44.590 --> 04:51.040 All right so this also works and it depends actually on the case but then you drop in the column so 04:51.070 --> 04:54.370 that you don't need or you select the columns that you need. 04:54.370 --> 05:00.760 So let's assume we have a data frame with 100 columns and you want to drop one column then you should 05:00.760 --> 05:03.020 definitely use the drop method. 05:03.010 --> 05:08.830 But if you have a data frame with 100 columns and you want to keep only two columns then it makes more 05:08.830 --> 05:13.140 sense to explicitly select the two columns. 05:13.160 --> 05:17.540 All right so now we are finished with the session and I hope to see you in the next session by.