WEBVTT 00:00.810 --> 00:02.410 Hello and welcome back to our cost. 00:02.410 --> 00:06.740 So in this session we are creating random numbers with the unpaid package. 00:07.050 --> 00:13.340 And we have already learned that we can create random random numbers with the random module of the standard 00:13.350 --> 00:14.880 python library. 00:14.940 --> 00:21.210 However creating random numbers for smart umpires more flexible and provides a lot more functionality. 00:21.210 --> 00:26.820 And it's also fast so so there's actually no reason to use the random module loft space and standard 00:26.820 --> 00:31.230 library even we can use the power for the name pi random package. 00:31.230 --> 00:40.370 So now let's first import pi and then what we would do you want to create a 10 random integers between 00:40.370 --> 00:47.180 1 and 101 excluding so let's press shift tab and you can see. 00:47.270 --> 00:54.120 So random dot Rand int returns integers font lo inclusive to high exclusive. 00:54.740 --> 01:03.260 So if you want random integers between 1 and 100 including we have to type year 101 into the third parameter 01:03.620 --> 01:04.520 is the size. 01:04.520 --> 01:12.890 So how many random numbers or integers we want to create let's execute the of here. 01:12.980 --> 01:20.210 Now you can see we get 10 random integers in the range one to one hundred and if we repeat this process 01:20.420 --> 01:29.660 we should get a different 10 random in the just because it's random and that's a very helpful instrument 01:29.660 --> 01:31.230 called random seed. 01:31.370 --> 01:38.620 So random that seed creates so-called absurd or random numbers and it enables us to reproduce the other 01:38.630 --> 01:39.710 random numbers we create. 01:39.710 --> 01:46.580 So whenever you use them the same seed as ISV in brackets if you have to define a number so here it's 01:46.580 --> 01:52.230 1 2 3 but can be any number and whenever you have we had the same number and the random seed to them 01:52.250 --> 01:56.630 you create exactly the same random numbers as I do. 01:56.920 --> 02:03.310 So let's run near the center and that's repeated the process and we should get the same numbers again 02:03.400 --> 02:05.680 because we didn't change for the random seed. 02:05.890 --> 02:07.030 All right. 02:07.150 --> 02:10.200 And we can also create normally distributed numbers. 02:10.210 --> 02:15.460 I don't want to go into details of normal distribution but it's kind of a better shape probability distribution 02:15.520 --> 02:18.340 where we can determine the means or minus 5. 02:18.880 --> 02:22.450 But it's flexible and you can determine the standard deviation. 02:22.480 --> 02:25.990 So the second position is 2. 02:25.990 --> 02:30.790 So Locke stands for the mean scale stands for the standard deviation and you can also determine the 02:30.790 --> 02:31.290 size. 02:31.300 --> 02:34.890 So how many normal distribution numbers you want to get. 02:34.900 --> 02:41.800 So when we executed we had 10 normal distribution numbers with mean 5 and standard deviation 2. 02:41.830 --> 02:50.170 All right so let's create an array with numbers between 1 and 100 so called B and then we can randomly 02:50.170 --> 02:57.880 shuffle all elements here so we can mix mix all elements up and change the sorting. 02:57.940 --> 03:02.800 Now we have to print B and you can see so be it. 03:02.800 --> 03:05.770 Now we are randomly sorted are not sorted. 03:05.770 --> 03:15.450 Actually but that's the method topside and where we can start again our unsorted array and let's check 03:15.450 --> 03:18.070 the results. 03:18.160 --> 03:24.220 Now it's sorted again now and actually for whatever reason so the method sort does not provide us the 03:24.220 --> 03:28.990 option to to sort the array in an inverse order or our reverse order. 03:28.990 --> 03:34.810 So I don't know why there's no option for it but yeah it's the case but we if you want to sort our array 03:34.810 --> 03:42.070 in the reverse order there's a trick here by double colon minus one where we have the reverse order 03:42.070 --> 03:48.100 starting from hundreds till 1 and that's another method called random that's why it's so random that 03:48.100 --> 03:55.330 choice creates a random sample from a given array so we can have our array B and then we determine the 03:55.330 --> 03:56.520 size of our sample. 03:56.510 --> 04:03.220 So for example one hundred and then we can determine whether the sample is with or without replacement. 04:03.280 --> 04:04.660 So what does that mean. 04:04.660 --> 04:10.810 Whenever an element is randomly chosen so let's say 1 the next step. 04:10.850 --> 04:13.730 Again we have the chance to choose this element. 04:13.730 --> 04:19.760 And with out replacement means that you can choose one element only once. 04:19.760 --> 04:27.740 So here we have a replacement is true and we choose a randomly select the sample of 100 elements from 04:27.740 --> 04:28.740 our array B. 04:29.510 --> 04:31.040 Let's see what we get. 04:31.040 --> 04:31.300 OK. 04:31.310 --> 04:40.020 We get a hundred elements and we can then swap our new sample array B1 and they can see so we have some 04:40.020 --> 04:48.360 duplicates so we have two was chosen two times four was chosen four times and with the method unique 04:48.830 --> 04:54.070 we can choose only the unique elements that we can remove all duplicates. 04:54.070 --> 05:01.010 There you can see we have only the unique values without any duplicates and you would get the same result 05:01.020 --> 05:06.780 by transforming our array into a set and then transforming it into a list and then to our right. 05:06.770 --> 05:10.420 So we cannot immediately transform a set into an array. 05:10.530 --> 05:15.800 So first we have to make a list and we get the same result here. 05:16.110 --> 05:22.640 And now by adding the method types you can determine how many unique elements we have here. 05:22.650 --> 05:29.160 You determine the unique elements of our sample and then that size gives us them the amount of unique 05:29.160 --> 05:32.410 values or elements of it 66. 05:32.430 --> 05:41.210 So we created 100 integers with replacement and we got back 66 unique elements and the unique method 05:41.210 --> 05:42.300 is quite informative. 05:42.300 --> 05:48.360 So it has some more parameters for example a return index it goes through and return count sequence 05:48.360 --> 05:50.310 true. 05:50.310 --> 05:53.020 So by default they are set to false. 05:53.060 --> 05:54.300 Let's see what we get. 05:54.510 --> 06:01.140 So here in the first array we have our unique elements and in the second array we get the index position 06:01.140 --> 06:05.010 of the first occurrence of our unique values. 06:05.010 --> 06:09.160 So for example if we go to our array with duplicates. 06:09.300 --> 06:17.130 The first occurrences to us of course at position 0 and the first occurrence of 3 is that position in 06:17.130 --> 06:26.110 x position to that I can see here and also kind of returns as an array with return counts. 06:26.160 --> 06:33.120 We can read that to us included in our sample two times three was included in our sample one time and 06:33.120 --> 06:37.010 four was then included in our sample four times. 06:37.080 --> 06:43.080 So this is quite informative here and we can also set the replace parameter to forwards here 06:49.230 --> 06:54.660 now there is actually no surprise because our array contains 100 elements and the free randomly select 06:54.750 --> 06:57.060 100 elements without replacement. 06:57.780 --> 07:04.440 How we actually get all 100 elements so check unique elements so we get all elements from 1 to 100 07:07.140 --> 07:10.890 and we have 100 unique elements. 07:10.890 --> 07:11.210 All right. 07:11.220 --> 07:16.520 So this was a short introduction to the number random package and we learned to create random integers 07:17.160 --> 07:20.880 normal distributed numbers and how to randomly shuffle an array. 07:20.910 --> 07:26.310 And also we learned how to make a random sample of an array and there are many more things you can do 07:26.310 --> 07:27.790 with the random module. 07:28.020 --> 07:34.180 So for example that's NPR random thought. 07:34.180 --> 07:37.970 And then if you press tab there you can see how many methods. 07:38.050 --> 07:38.560 You can. 07:38.560 --> 07:43.570 You can apply here so you can have an anomalous simulation. 07:43.570 --> 07:50.950 And yes many many things you can ask the price and distribution and so on and so on. 07:50.950 --> 07:51.300 All right. 07:51.310 --> 07:57.370 So this is it for the time being so this was a short intro to random numbers and the next session we 07:57.370 --> 08:03.330 will compare the python standard library with the PI packets in terms of performance so hope to see 08:03.330 --> 08:04.900 you there and by.