WEBVTT 00:00.680 --> 00:06.290 In this video we learn how to add a brand new column to our data frame so later and the cost. 00:06.290 --> 00:12.590 And there's a complete section about concatenate emerging and joining off data friends and as areas. 00:12.590 --> 00:17.710 So this video is meant to be a short introduction to this topic. 00:17.740 --> 00:26.290 All right so let's first import pandas and we also import our Titanic's dataset and we have a look at 00:26.290 --> 00:32.940 the first five rows and now we want to create a brand new column with only one value so all of those 00:32.940 --> 00:39.420 have the same value and we can do this with the so-called broadcasting. 00:39.420 --> 00:43.740 And let's assume that the column should only have zeros on all rows as a zero. 00:44.160 --> 00:51.380 And we want to call our columns the rows so the labels should be zeros and what we first can do what 00:51.380 --> 00:58.070 we can actually select a column by passing the column in quotation marks within this span brackets here 00:58.670 --> 01:05.130 and let's check if we have already a column zero so I would guess that we get an error here. 01:06.660 --> 01:08.940 So there is no zeros as column label. 01:10.050 --> 01:15.910 And what we can do now we can create a new column called zeroes and we can assign a value. 01:15.960 --> 01:18.180 And that's zero here. 01:18.270 --> 01:20.700 So let's see what we get. 01:20.700 --> 01:22.790 So let's have a look here at our data frame. 01:24.670 --> 01:32.390 And here we can see on the right side of that there is a new column with the label zeros and the values 01:32.480 --> 01:39.130 0 and it's actually not necessary to have a integer offload as new value. 01:39.130 --> 01:45.640 We can also pass here a string 0 and let's check again. 01:45.640 --> 01:53.590 So now we overwrite the existing is that row column here and now we have our and use the row column 01:53.710 --> 02:01.390 with them the values the row as a string before we learn that we can select a single column either by 02:01.390 --> 02:11.530 the spec records notation here or by calling the attribute here and both notations are when selecting 02:11.530 --> 02:13.300 single columns are the same. 02:13.360 --> 02:17.640 And now it comes at the most important difference between those two notations. 02:18.640 --> 02:24.850 So with the span brackets notation b not only can select an existing column we can also create a new 02:24.850 --> 02:27.580 column as you see here. 02:27.730 --> 02:35.040 However with them the attribute notation we only can select existing columns so we cannot create new 02:35.040 --> 02:43.690 columns however we can try to do so and we could try to create a column once with the one as a value 02:45.590 --> 02:50.180 and we can have a look at our data frame. 02:50.180 --> 02:58.920 And obviously there is no column with the label ones and the values one however we can call the attribute 02:58.920 --> 03:03.650 1 so maybe something is uh saved here. 03:03.730 --> 03:09.630 Yeah and actually we created the attribute once and that's just a single value of one. 03:10.620 --> 03:16.470 So far the class Panda's data frame we created an attribute once and we assign the value one. 03:16.530 --> 03:21.080 But actually we didn't create a brand new column. 03:21.140 --> 03:21.590 All right. 03:21.590 --> 03:25.790 So now we are finished with this session and I hope to see you in the next session by.