1 00:00:00,530 --> 00:00:04,390 All right now I know in the last lecture we said well we're going to I'm going to jump into two different 2 00:00:04,390 --> 00:00:04,750 tools. 3 00:00:04,750 --> 00:00:10,510 But I thought if we get into the next section I just want to really highlight that all the steps in 4 00:00:10,510 --> 00:00:16,540 this process won't be possible without the right tools and the be a detailed section on each of these. 5 00:00:16,600 --> 00:00:18,690 But it's a little sneaky preview. 6 00:00:19,240 --> 00:00:21,790 Let's have a look at what we'll be using. 7 00:00:21,790 --> 00:00:27,610 The first thing is getting your computer set up and ready to use tools that can be used for these different 8 00:00:27,610 --> 00:00:28,680 steps. 9 00:00:28,690 --> 00:00:34,600 Think of it as creating a work shed for Data Science and machine learning projects on your actual computer. 10 00:00:35,660 --> 00:00:42,200 To do this we'll use Anaconda which is like the hardware store of data science and machine learning 11 00:00:42,200 --> 00:00:42,950 tools. 12 00:00:42,950 --> 00:00:44,150 So it all starts in the base. 13 00:00:44,150 --> 00:00:45,660 We've got your computer here. 14 00:00:45,740 --> 00:00:52,130 We're going to install Anaconda and then once you've got Anaconda getting all of these other machine 15 00:00:52,130 --> 00:00:59,180 learning tools like Jupiter notebooks pi torch name pi map plot layer pandas psyche can't loan cat boost 16 00:00:59,420 --> 00:01:07,220 ex G boost and tensor flow well some of these are a mouthful but once you've got Anaconda getting these 17 00:01:07,220 --> 00:01:13,240 tools that we need to to apply to that process we've just created that framework we've created will 18 00:01:13,250 --> 00:01:14,990 be an absolute breeze. 19 00:01:15,020 --> 00:01:17,600 So you've probably got these at the moment your computer. 20 00:01:17,720 --> 00:01:22,640 Next we'll have a detailed section on on most of these each of these because there's a lot of machine 21 00:01:22,640 --> 00:01:23,500 learning tools out there. 22 00:01:23,510 --> 00:01:27,770 But I want to make sure that you've got hands on knowledge with some practical ones that you can get 23 00:01:27,770 --> 00:01:29,030 started using straight away. 24 00:01:29,360 --> 00:01:33,350 So what will this look like if we wanted to attach it to our process. 25 00:01:33,680 --> 00:01:40,850 Well to write Python code and communicating our work we're going to be using Jupiter notebooks and these 26 00:01:40,850 --> 00:01:45,580 are also the same kind of file you'll find all of the projects are done in. 27 00:01:45,950 --> 00:01:50,390 If you stick with it you're going to finish the course with a few of these containing real world machine 28 00:01:50,390 --> 00:01:55,250 learning projects where you can show people what you've done for data analysis. 29 00:01:55,250 --> 00:01:59,700 So this little section here and we've we've done some color coding here on this diagram. 30 00:01:59,810 --> 00:02:01,480 We've got this whole whole thing here. 31 00:02:01,490 --> 00:02:06,860 This whole modeling section of problem definition experiments and data analysis machine learning that 32 00:02:06,860 --> 00:02:08,770 can be thought of as data science. 33 00:02:08,780 --> 00:02:16,040 We've seen a similar graphic to this in the past but this is now mapping the tools which are these two 34 00:02:16,040 --> 00:02:18,620 hour framework that we've designed before. 35 00:02:18,620 --> 00:02:24,500 So for data analysis you'll be using things like pandas that plot lib and num pi. 36 00:02:24,770 --> 00:02:30,920 That's step two three and four building machine learning models you'll use things like tensor flow pi 37 00:02:30,930 --> 00:02:35,630 torch psychic learn ex G boost and cat boost. 38 00:02:35,630 --> 00:02:41,300 And what's important to remember is your most important role will not be knowing all of the functions 39 00:02:41,300 --> 00:02:44,060 of each of these single libraries off by heart. 40 00:02:44,060 --> 00:02:48,750 It will be knowing which tool to use for what kind of problem. 41 00:02:48,860 --> 00:02:55,100 Again that's why Step 1 is defining our problem so you can know I can draw out of my my tool bag here 42 00:02:55,100 --> 00:03:00,490 which which is this I can apply these tools are these stamps and these tools for these stamps and it's 43 00:03:00,490 --> 00:03:03,680 gonna be all done in Anaconda and Jupiter notebooks. 44 00:03:03,980 --> 00:03:05,680 Sometimes you won't know the answer. 45 00:03:05,930 --> 00:03:11,630 Sometimes I don't even know the answer but since I've had some practice I know where to look I've kind 46 00:03:11,630 --> 00:03:15,320 of thought about these different things I've had my hands on practice with these different things so 47 00:03:15,320 --> 00:03:16,960 I know where to come to look. 48 00:03:17,450 --> 00:03:19,180 And eventually you will too. 49 00:03:19,190 --> 00:03:20,560 That's what we're focused on. 50 00:03:21,610 --> 00:03:21,880 Okay. 51 00:03:21,920 --> 00:03:22,990 Okay that's enough. 52 00:03:23,000 --> 00:03:28,050 You should be proud you've learned some of the most fundamental concepts in machine learning. 53 00:03:28,100 --> 00:03:32,180 Of course there are many more and there's there's different ways to do the same thing. 54 00:03:32,540 --> 00:03:39,170 But these are more than enough what we've covered so far to start becoming a machine learning practitioner. 55 00:03:39,170 --> 00:03:45,650 Up next we're going to get Anaconda installed on your computer because if we look at the diagram before 56 00:03:46,970 --> 00:03:55,610 everything starts with Anaconda once you have your computer you install Anaconda you get these tools. 57 00:03:55,610 --> 00:03:59,090 So that's what we're going to do next let's do it.