1 00:00:00,750 --> 00:00:07,770 Hello and welcome to the very first practical tutorial for this course and this is from this point that 2 00:00:07,770 --> 00:00:14,670 we're going to build this very powerful but very simple model an artificial intelligence which I'm still 3 00:00:14,670 --> 00:00:19,100 blown away to see the insane results it is able to perform. 4 00:00:19,170 --> 00:00:22,410 I remind that this mall was made just recently. 5 00:00:22,410 --> 00:00:29,970 You know the research paper appeared in March in late March 2018 and it surpasses any other AI algorithm 6 00:00:30,180 --> 00:00:32,910 that we've worked on in the year 2017. 7 00:00:32,910 --> 00:00:34,540 So it is a big deal. 8 00:00:34,560 --> 00:00:39,570 What we're about to implement and the double good news is that we don't need any library. 9 00:00:39,570 --> 00:00:45,130 You know it manages to beat the AI models we built on tend of flow or by torch. 10 00:00:45,150 --> 00:00:51,850 It's crazy it's just pure logic and that's why we will be able to implement it with only the non-pay 11 00:00:51,850 --> 00:00:52,730 library. 12 00:00:52,770 --> 00:01:00,950 We will not need any sense of flow or pie torch or any sort of advanced AI framework we will just need 13 00:01:00,950 --> 00:01:01,780 them by. 14 00:01:01,800 --> 00:01:04,620 You will see the results will be insane. 15 00:01:04,620 --> 00:01:05,260 All right. 16 00:01:05,340 --> 00:01:09,720 But let's start by installing an icon that's why right now I'm on Google. 17 00:01:09,720 --> 00:01:10,540 Let's dive here. 18 00:01:10,590 --> 00:01:14,010 Anaconda you can just type panic on that. 19 00:01:14,080 --> 00:01:16,800 Will get you the first link instead of the steak. 20 00:01:16,960 --> 00:01:17,600 Here we go. 21 00:01:17,620 --> 00:01:18,990 And on the cloud. 22 00:01:19,270 --> 00:01:23,460 And now you just need to go here to download Anaconda. 23 00:01:23,770 --> 00:01:28,240 And here we go now choose Anaconda for the system you're having. 24 00:01:28,240 --> 00:01:30,210 Windows Mac OS or Linux. 25 00:01:30,400 --> 00:01:38,100 I'm a Mac OS so I'm going to write and make sure to download five 3.6 instead of pipe and 2.7. 26 00:01:38,180 --> 00:01:38,440 Right. 27 00:01:38,440 --> 00:01:43,600 So if you're ready let's click on download then we'll get the file and we will just install an icon 28 00:01:43,600 --> 00:01:49,260 that will just take one or two minutes and then we'll be ready to start the implementation. 29 00:01:49,660 --> 00:01:51,490 All right so two seconds left. 30 00:01:51,490 --> 00:01:52,340 Here we go. 31 00:01:53,050 --> 00:01:54,760 Perfect We have an account. 32 00:01:55,060 --> 00:01:59,500 So now I'm going to jump to my downloads folder here it is. 33 00:01:59,500 --> 00:02:03,100 And here is the file I've just downloaded. 34 00:02:03,100 --> 00:02:04,650 Let's double click on it. 35 00:02:04,720 --> 00:02:08,780 Here you just need to click on continue continue continue continue. 36 00:02:09,070 --> 00:02:12,220 You can read if you want and agree and install. 37 00:02:12,220 --> 00:02:13,210 And there we go. 38 00:02:13,210 --> 00:02:14,780 The installation is on its way. 39 00:02:14,810 --> 00:02:20,410 We'll just take a few seconds or maybe one or two minutes but not more. 40 00:02:20,690 --> 00:02:27,190 So while it's installing I can tell you a few words about Anaconda and I can do is basically a platform 41 00:02:27,280 --> 00:02:33,820 that contains several IDs integrated development environments on which you can code and program anything 42 00:02:33,820 --> 00:02:40,390 you want doesn't have to be any IDE can be you know some basic tools and deterministic tools but of 43 00:02:40,390 --> 00:02:45,270 course on one of those frameworks and one of those ideas will build an AI. 44 00:02:45,490 --> 00:02:50,330 And this idea will be spider as in every of our courses. 45 00:02:50,380 --> 00:02:56,980 Spider is the best for me at least when you're beginning or when you need a nice interface because indeed 46 00:02:57,280 --> 00:03:01,860 it's really like you're in a studio and you can visualize the data easily. 47 00:03:01,900 --> 00:03:06,330 You can open some files from the file explorer you can debug easily. 48 00:03:06,470 --> 00:03:10,340 Anyway it's a great tool to implement anything you want. 49 00:03:10,340 --> 00:03:11,500 So we'll go for spider. 50 00:03:11,530 --> 00:03:13,420 I'll show you Spider very soon. 51 00:03:13,420 --> 00:03:13,810 Here we go. 52 00:03:13,810 --> 00:03:15,430 The installation is over. 53 00:03:15,460 --> 00:03:16,920 You don't have to install Microsoft. 54 00:03:16,930 --> 00:03:23,530 Yes code you can just click Continue close and you can choose to keep the file the downloading felt 55 00:03:23,560 --> 00:03:25,450 or not perfect. 56 00:03:25,450 --> 00:03:32,500 So now for those of you who are on Windows go at the bottom left of your monitor to go into the list 57 00:03:32,500 --> 00:03:34,950 of program and find Anaconda. 58 00:03:35,380 --> 00:03:40,100 For those of you who are on Linux you will find an icon then same in the list of programs. 59 00:03:40,210 --> 00:03:46,540 And for those of you who are on Mac like me you just need to get applications and double click on this 60 00:03:46,930 --> 00:03:48,820 anaconda Navigator. 61 00:03:48,820 --> 00:03:54,840 It will open this whole Anaconda framework that contains several ideas. 62 00:03:54,940 --> 00:03:56,170 I will show you the ideas. 63 00:03:56,170 --> 00:03:57,230 Here they are. 64 00:03:57,240 --> 00:03:59,740 So Jupiter notebook is one ID. 65 00:03:59,740 --> 00:04:05,670 One great idea actually if you want to combine you know programming and mathematics written in lasic 66 00:04:05,680 --> 00:04:11,980 for example if you're writing a research paper that combines some code in math and you have spider as 67 00:04:11,980 --> 00:04:18,230 I told you about you also have our studio if you want to code in are you can code in our any. 68 00:04:18,280 --> 00:04:23,460 That's totally possible or some deep variables but I mostly use are for statistics. 69 00:04:23,540 --> 00:04:28,270 And now as far as we are concerned we are going to work on spiders. 70 00:04:28,270 --> 00:04:31,240 You just need to click on launch and there we go. 71 00:04:31,240 --> 00:04:36,750 Spider is opening sending up main window and here you go. 72 00:04:36,750 --> 00:04:38,730 Welcome to Spider. 73 00:04:38,730 --> 00:04:41,600 This is the spider ID interface. 74 00:04:42,000 --> 00:04:47,200 So the first thing you need to understand when you open Spier is that you'll get this temporary tender 75 00:04:47,370 --> 00:04:50,990 wife out and we will not be using that file at all. 76 00:04:51,000 --> 00:04:54,700 It's just a temporary felt like there has to be a fall in here. 77 00:04:54,810 --> 00:05:01,550 That's the editor that a window which can combine File Explorer will you have all your files. 78 00:05:01,650 --> 00:05:07,890 Variable Explorer which will contain all the variables you create and help which can give you some information 79 00:05:07,890 --> 00:05:14,850 on the different objects and classes that you're using while building any I remind that for our age 80 00:05:14,860 --> 00:05:21,190 I will not be using any library so we will probably not use the help but it's very helpful. 81 00:05:21,510 --> 00:05:21,930 All right. 82 00:05:21,930 --> 00:05:28,650 And then this is the Council meaning that when we execute our code our code will be running inside this 83 00:05:28,650 --> 00:05:34,140 console and will execute the training of the AI for what we're going to do. 84 00:05:34,230 --> 00:05:34,620 Perfect. 85 00:05:34,620 --> 00:05:41,520 So that spider and now the first thing we're going to do is go to falx poor and I'm going to go to my 86 00:05:41,730 --> 00:05:49,780 desktop because in my desktop I created this 8:28 in folder that will contain all the code will be implementing 87 00:05:49,810 --> 00:05:50,870 in this course. 88 00:05:51,000 --> 00:05:56,100 Right now this folder is empty but it is this folder that will contain the file that we're about to 89 00:05:56,490 --> 00:06:00,030 create and which will contain the whole artificial intelligence. 90 00:06:00,030 --> 00:06:05,820 We will build in this course we will actually build several ones but we will start with the main one 91 00:06:06,150 --> 00:06:12,780 that will contain the heart of the artificial intelligence following exactly this new research paper 92 00:06:12,780 --> 00:06:20,160 that just came out in 2018 in March 2018 and then we'll connect the AI to the environment to train it 93 00:06:20,400 --> 00:06:22,470 to accomplish what we want to do. 94 00:06:22,470 --> 00:06:23,520 I'll get to there soon. 95 00:06:23,520 --> 00:06:28,320 But we will build the AI here and our file will be saved in this folder. 96 00:06:28,380 --> 00:06:29,250 I. 97 00:06:29,670 --> 00:06:31,120 So feel free to do the same. 98 00:06:31,140 --> 00:06:35,020 And if that's the case you're ready to click on file at the top. 99 00:06:35,100 --> 00:06:36,580 Click on new file. 100 00:06:36,600 --> 00:06:40,050 This will be the pipe and file this will contain the AI implementations. 101 00:06:40,050 --> 00:06:41,680 Now you can close that. 102 00:06:41,850 --> 00:06:47,790 And now you can select all this delete then added Tyrel we're going to get to this implementation the 103 00:06:47,790 --> 00:06:49,000 following title. 104 00:06:49,020 --> 00:06:54,770 So you have to put a hash tag first because the title has to be in a comment hash tag. 105 00:06:54,870 --> 00:06:58,040 And then let's just call it a 20:18. 106 00:06:58,050 --> 00:06:58,490 All right. 107 00:06:58,500 --> 00:07:03,680 We're going to build the AI 20:18 the most powerful AI 2018. 108 00:07:03,780 --> 00:07:09,360 And actually I think it is the most powerful AI AI mole ever built because you're going to see how it 109 00:07:09,630 --> 00:07:14,270 surpasses any score at least on the environment that we're going to try. 110 00:07:14,730 --> 00:07:15,220 All right. 111 00:07:15,270 --> 00:07:16,550 So that's the title. 112 00:07:16,560 --> 00:07:19,520 Let's just start with this and then we're going to save. 113 00:07:19,680 --> 00:07:28,200 And we can just call it a r s because indeed the ideas we're going to build is called commented random 114 00:07:28,500 --> 00:07:29,420 search. 115 00:07:29,490 --> 00:07:30,820 So let's save that. 116 00:07:30,870 --> 00:07:31,660 And here we go. 117 00:07:31,680 --> 00:07:34,510 We are ready to start the Egmond and random search. 118 00:07:34,620 --> 00:07:39,650 So we'll stop here for this to toile then in the next one hour introduce you to the environment. 119 00:07:39,760 --> 00:07:44,540 I'll introduce you to the research paper Kiril explain the whole intuition behind it. 120 00:07:44,730 --> 00:07:50,250 But since you know we're going to go back and forth jumping from the research paper into the code because 121 00:07:50,250 --> 00:07:55,200 I want to show you exactly what we're coding with respect to the research paper. 122 00:07:55,320 --> 00:08:00,410 I prefer to already introduce it to you so that we can already get familiar with it. 123 00:08:00,630 --> 00:08:01,060 Perfect. 124 00:08:01,060 --> 00:08:03,040 So I'll see you in the next tutorial. 125 00:08:03,060 --> 00:08:04,830 And until then enjoy AI.