WEBVTT 00:01.510 --> 00:02.620 Hi and welcome back. 00:02.620 --> 00:07.670 So in this video we continue our first inspection of our Titanic dataset. 00:07.900 --> 00:11.200 So in the last video we print out our whole dataset. 00:11.800 --> 00:17.080 However with the large dataset with many rows it doesn't make sense to print the whole dataset. 00:17.530 --> 00:24.180 And in most cases to get a very first good impression it's enough to have a look at the first five rows. 00:24.190 --> 00:32.050 And the last five rows for example and therefore there are two very hard four methods here. 00:32.050 --> 00:41.230 So first of all we have the head method and that's press shift type here. 00:41.410 --> 00:48.110 So the happy method returns to the first end rows of our data frame and by default the parameter end 00:48.120 --> 00:49.060 is set to five. 00:49.060 --> 00:54.970 But yeah we can change it but now for the time being let's try it here with them the default parameter 00:56.060 --> 01:03.380 and we can see here now our very first five rows of our data frame and we can pass different values 01:03.380 --> 01:04.210 to our parameter. 01:04.220 --> 01:12.830 And so for example if you want to have a look at the first 10 rows we can set end to 10 and then we 01:12.830 --> 01:14.430 get the first 10 rows. 01:14.480 --> 01:21.710 If it's enough to get only the first two rows and then we can pass to. 01:21.800 --> 01:28.040 So now we know the very first rows and in combination with the the very last draws of our data frame. 01:28.070 --> 01:36.020 So this typically gives us a very good overview over the dataset and we can call the very last throws 01:36.050 --> 01:37.730 this then the Taylor method 01:42.640 --> 01:45.740 so by default the tail method returns. 01:45.790 --> 01:50.550 The last end rows and NASA set by default to 5. 01:50.560 --> 01:57.460 But we can change also this year and let's say we want to have the last 10 rows and we got the last 01:57.460 --> 02:05.220 10 rows and you can see here with the help method and also the tail method we can see here the column 02:05.220 --> 02:09.270 labels and we can also explicitly call our column labels 02:12.130 --> 02:14.350 with the attribute columns 02:17.910 --> 02:20.080 and we get a so-called index object. 02:20.100 --> 02:27.780 So it looks like a list of this on the column labeled slow survived P Class 6 it's the best P Podge 02:27.790 --> 02:35.780 fair embarked Imtech and we can also explicitly call our row labels here on the left side 02:39.020 --> 02:41.540 with the attribute top index 02:44.450 --> 02:49.750 and we can see that we have a so-called Range Index so starting from zero until eight hundred ninety 02:49.750 --> 02:57.480 one and eight hundred and ninety one excluding so we was started with a zero until eight hundred ninety 02:57.950 --> 03:06.060 and this is a so-called Range Index and that's a very helpful method called in for where we can get 03:06.060 --> 03:16.310 made our data on our data frame so let's try this one here so that in for and here we get meta information 03:16.310 --> 03:22.360 on our data frame so for example here you can see that it's a data frame and that we have a Range Index 03:22.370 --> 03:32.640 with 891 entries from zero to 890 and we see that we have in total nine columns and here we have them 03:32.650 --> 03:39.070 the labels of the columns so survived the passenger class onto the deck and here we can see the amount 03:39.070 --> 03:45.430 of non null null values so typically in the data set they are missing value as are so called No values 03:46.460 --> 03:49.930 And here we get the information how many are known missing values we have. 03:49.930 --> 03:56.950 So here in these survived passenger class and sex column we do not have any missing value so we have 03:57.070 --> 04:05.310 values for each row and for the H column here we have obviously one hundred seventy seven missing values 04:05.310 --> 04:14.230 so far 177 passengers we actually do not know the H and in the deck column there are obviously many 04:14.230 --> 04:22.380 missing values and here on the right side you can see that data type in each column so in the survive 04:22.380 --> 04:28.500 column B a half either serial or one it's an integer an awesome the passenger class column we have one 04:28.510 --> 04:29.740 two 3. 04:29.850 --> 04:37.920 This is also an integer then in the sixth column we have object and in very easy words object stands 04:37.920 --> 04:43.640 for a string data type and then we have H which is a float. 04:43.650 --> 04:48.150 So for example a thirty three point zero years 04:50.860 --> 04:54.180 and in this sub S P and patch columns we have an integer. 04:54.190 --> 05:01.450 So the amount of siblings for parents and for in the fact column again we have a float. 05:01.450 --> 05:09.900 So the price for the ticket that the peasants are paid for example is seven point eight nine and embarked 05:09.900 --> 05:10.450 on deck. 05:10.450 --> 05:16.680 These are both columns with a string objects actually here so you can see S and B 05:20.440 --> 05:24.940 and then you can see a summary of the data types so we have two columns with float objects. 05:24.940 --> 05:34.170 We have four columns in the US and we have three columns with strings and the total memory usage of 05:34.170 --> 05:37.270 our data frame is sixty two point seven kilobytes. 05:37.290 --> 05:40.830 So we have here a very small dataset actually. 05:40.950 --> 05:44.280 So that data sets its megabytes and gigabytes. 05:44.910 --> 05:47.240 So this is very small and easy to handle. 05:48.600 --> 05:52.770 And that's also a attribute called shape. 05:54.720 --> 05:59.700 So this does not actually give us more information but it gets that's in the shape of our data frame. 05:59.700 --> 06:08.150 We have eight hundred ninety one rows and nine columns and last but not least that's the very helpful 06:08.150 --> 06:12.250 method to describe thought. 06:12.270 --> 06:13.100 Describe 06:18.890 --> 06:21.710 and here we can see for on American columns. 06:21.710 --> 06:29.980 So we have here integer and float we have 1 2 3 4 5 6 numerical columns. 06:29.990 --> 06:36.020 We can see here a summary for each columns then yeah actually a summary statistics. 06:36.110 --> 06:42.840 So column stands for the amount of num low values and in this survival column we have eight hundred 06:42.840 --> 06:44.360 ninety one num low values. 06:44.360 --> 06:51.560 So these are actually here the same values as here and we can see here in the column that we have missing 06:51.560 --> 06:55.250 values then next. 06:55.300 --> 06:58.350 Here we f and the mean for all columns. 06:58.720 --> 07:04.050 And we can see here that in total thirty eight percent of all passengers only survived. 07:04.150 --> 07:14.390 So 62 percent um died in the Titanic disaster and the Emirates passenger class is two point three. 07:14.390 --> 07:19.690 So they are much more passengers and class the two and three than in class one. 07:20.120 --> 07:30.240 And the average age is twenty nine and the average ticket price was a 30 to then we have here the standard 07:30.240 --> 07:37.140 deviation for all columns and the minimum so we can see here the youngest passenger was 0 point for 07:37.140 --> 07:44.110 2 years old um actually a baby and there were passengers that died. 07:44.120 --> 07:52.660 So the minimum value is 0 0 4 died and one four survived and with the passenger classes one two three 07:52.670 --> 08:00.000 the minimum is one year and we had actually passengers uh that uh didn't pay anything for that ticket 08:01.120 --> 08:07.480 and then here we have the 25th percentile we have the fifth person tile which is actually the median 08:08.230 --> 08:16.220 and we have uh seventy first person tile and we have the maximum for each column so there are passengers 08:16.220 --> 08:23.840 who survived um the last passenger class this class 3 the oldest passenger on board was 80 years old 08:25.210 --> 08:31.960 and that was the passenger that had eight siblings and spouses aboard and that was the presence of US 08:31.970 --> 08:40.820 the six children or parents onboard of the Titanic and the highest share that was paid was five hundred 08:40.820 --> 08:42.500 twelve dollar a pound. 08:42.500 --> 08:45.820 So obviously a presence in class one. 08:45.900 --> 08:51.360 So now we are finished with the first inspection of our dataset and only after a few minutes and a few 08:51.360 --> 08:57.750 lines of code we actually know a lot of information about our data set here and in the next sessions 08:57.780 --> 09:02.080 we have further build on our skills to further analyze our data set. 09:02.100 --> 09:03.870 Yes so hope to see you there by.