WEBVTT 00:01.710 --> 00:07.460 In this video we learn how to filter data frames based on many conditions so that must all be met. 00:07.470 --> 00:13.530 So first of all we import pandas and we are working with our Titanic data frame. 00:13.740 --> 00:22.040 And here we have the first 10 rows and from the movie we know that after hitting the iceberg there was 00:22.040 --> 00:26.740 a clear code of conduct to first save all women and children. 00:26.870 --> 00:33.830 So women and children had priority for the few lifeboats that were onboard and therefore we can make 00:33.830 --> 00:40.160 the hypothesis that being an adult man was a pretty bad combination for surviving the disaster. 00:40.390 --> 00:47.580 And yeah actually what do you want to do we want to check now if the data supports our hypothesis so 00:47.580 --> 00:51.130 we want to filter all mail that were older than 14 years old. 00:51.150 --> 00:57.970 So you see a 14 years but we could also use 69 in years and actually we have two conditions. 00:58.230 --> 01:06.420 So the passenger must be a male so titanic not sex equals male and we started the boolean serious and 01:06.500 --> 01:13.980 mask one variable and the second condition is that the H must be greater than 14 and we save for the 01:13.980 --> 01:17.610 reciting boolean series in the variable mask too. 01:17.700 --> 01:25.510 So let's check here and we check the first five elements of each mask. 01:25.560 --> 01:31.860 So now we have to combine both conditions with an end operator and only if both conditions are met we 01:31.860 --> 01:33.460 get a true. 01:33.750 --> 01:42.020 So for being an adult male the person must be a male and it must be older than 14 years old so we can 01:42.020 --> 01:48.080 hear use the end sign and combine mask one and Basque two with the end sign and we will have a look 01:48.080 --> 01:50.860 at the first five elements. 01:51.110 --> 01:53.110 So let's go to index Position ZERO. 01:53.120 --> 01:54.670 Here we have true and true. 01:54.700 --> 02:02.390 It's true then in exposition 1 we are false and true gives the false because both conditions must be 02:02.390 --> 02:10.940 met and that in x position 2 we have false and true again and we gather files and so on and now we can 02:10.940 --> 02:19.960 pass mask 1 and mask 2 into our local operator and by doing so we can filter out all the rows we have 02:19.970 --> 02:25.780 we have a male passenger that is older than 14 years old and we can also select here for the columns 02:26.240 --> 02:31.970 so we are only interested in these surveys column in the test was our class column in the sex column 02:31.970 --> 02:40.020 and in the age column so we create here a brand new data frame and we store the data frame in the variable 02:40.050 --> 02:46.940 male survived so let's run the other cell and check the first five rows of our new data frame 02:49.560 --> 02:56.160 and here we have our new data frame we have we have only made that older than 14 years so so we can 02:56.160 --> 02:58.980 also check that say 20 rows 03:04.180 --> 03:08.470 so only male passengers and only made that are older than 14 years old. 03:08.480 --> 03:13.980 So this works here. 03:14.150 --> 03:14.510 All right. 03:14.510 --> 03:19.040 So now we can also get more information on our new data frame with the info method 03:21.690 --> 03:31.840 so we have total 414 passengers that are male and older than 14 years old and let's use the described 03:31.840 --> 03:32.290 method 03:35.190 --> 03:44.810 so we have 414 passengers and we actually know that in the survival column we have 1 4 survived and 03:44.810 --> 03:53.920 the 0 4 not so life and we have a mean year of 0.01 7 in our survival column this means that only 17 03:53.920 --> 04:01.270 percent of all male passengers that were older than 14 years old actually survived the Titanic disaster 04:06.150 --> 04:09.440 and let's recall the survival rate for all passengers. 04:09.450 --> 04:18.060 So let's use the described method on the total data set here and here we can see that on total a thirty 04:18.060 --> 04:25.140 eight point three percent of all passengers survived but in the subset of male passengers older than 04:25.140 --> 04:25.990 14 years old. 04:26.010 --> 04:28.600 Only 17 percent survived. 04:28.860 --> 04:34.320 So it's pretty likely that our hypothesis is true that being an adult male passenger on board of the 04:34.320 --> 04:40.080 Titanic was a pretty bad combination for surviving the disaster. 04:40.110 --> 04:44.910 All right so now we are finished with this session and I hope to see you run next session by.