1 00:00:11,700 --> 00:00:16,920 In this lecture we are going to go over a very new and exciting environment for writing deep learning 2 00:00:16,920 --> 00:00:22,230 code in Python which is Google's core lab short for code laboratory. 3 00:00:22,230 --> 00:00:26,930 For those of you who like to use Jupiter notebook this is an even better option. 4 00:00:26,940 --> 00:00:33,480 This is basically the same as you put a notebook with the following bonuses first. 5 00:00:33,490 --> 00:00:37,520 It's hosted by Google which means you don't have to use your own computing power. 6 00:00:37,540 --> 00:00:43,330 You'll notice that when you need to download data files it happens extremely quickly because while Google's 7 00:00:43,330 --> 00:00:45,760 network is extremely fast. 8 00:00:45,760 --> 00:00:52,750 Second you get access to a GP you and even Google's new TB you which is pretty amazing a tepee you is 9 00:00:52,750 --> 00:00:55,560 not something you can buy for your personal computer. 10 00:00:55,780 --> 00:00:58,660 So it's pretty nice to be able to make use of one. 11 00:00:58,990 --> 00:01:03,670 Remember that the way pi torch code is written you don't have to worry about what kind of device you're 12 00:01:03,670 --> 00:01:04,300 using. 13 00:01:04,330 --> 00:01:06,960 Well for the most part TB you is another story. 14 00:01:07,180 --> 00:01:14,200 But generally speaking the same code will work whether you're using a CPA you or a GP you third the 15 00:01:14,200 --> 00:01:16,680 code lab notebooks are stored in your google drive. 16 00:01:16,690 --> 00:01:17,980 So it's in the cloud. 17 00:01:18,160 --> 00:01:22,210 You'll never lose it and it's very easy to share with other people. 18 00:01:22,210 --> 00:01:26,950 Fourth is that many of the libraries you need for a deep learning machine learning and data science 19 00:01:27,190 --> 00:01:28,950 are already included. 20 00:01:28,990 --> 00:01:32,860 In fact I was surprised that there were many more than I assumed there would be. 21 00:01:32,860 --> 00:01:38,660 There are even competing Deep Learning libraries already included such as piano and Pi torch. 22 00:01:38,680 --> 00:01:44,640 So for those of you who hate doing environments set up myself included this is really truly awesome. 23 00:01:49,730 --> 00:01:53,920 So in this lecture we're not going to do anything really technically complicated. 24 00:01:53,930 --> 00:01:57,850 Rather we're just going to talk about Google collab and do some short demos. 25 00:01:57,860 --> 00:02:04,100 So you know how it works and can see for yourself that it's just like writing Python anywhere else to 26 00:02:04,100 --> 00:02:04,550 start. 27 00:02:04,550 --> 00:02:10,550 I'm going to assume you already know how to create a Google Drive account if you don't have one go to 28 00:02:10,580 --> 00:02:13,100 drive dot google dot com and sign up 29 00:02:18,440 --> 00:02:23,890 once you have your google drive account and you've logged in you'll see this interface from here. 30 00:02:23,900 --> 00:02:29,930 You can hit the new menu which allows you to create all different kinds of files such as Google Docs 31 00:02:30,160 --> 00:02:32,690 a spreadsheet a presentation and so forth. 32 00:02:34,340 --> 00:02:35,710 So let's do this. 33 00:02:38,240 --> 00:02:49,480 So now what you want to do is go to the more menu and hit code laboratory. 34 00:02:49,570 --> 00:02:49,830 All right. 35 00:02:49,830 --> 00:02:52,380 So as you can see this brings up a new notebook. 36 00:02:52,530 --> 00:03:01,510 And from here you can mostly use this as you would a normal notebook. 37 00:03:01,530 --> 00:03:06,180 Now one thing that might happen to you is that you might not see cool laboratory in the menu at all. 38 00:03:06,780 --> 00:03:13,200 So as you can see I've hit the new menu and I've hit more but I don't see color in this case. 39 00:03:13,200 --> 00:03:14,790 Here's what you can do. 40 00:03:14,790 --> 00:03:19,600 You want to select connect more apps from here. 41 00:03:19,610 --> 00:03:21,050 Just search for collab. 42 00:03:21,200 --> 00:03:28,030 And the first thing that will pop up is Google's collab add this and Google Ecolab will become available 43 00:03:28,270 --> 00:03:28,920 from the menu. 44 00:03:28,930 --> 00:03:29,680 We just looked at 45 00:03:37,260 --> 00:03:41,190 so if we go there again we can see that cold app and now appears where it should 46 00:03:48,670 --> 00:03:59,930 so let's go in and rename this notebook to TAF 2.0 intro. 47 00:03:59,960 --> 00:04:01,810 So first we're gonna get right to the good stuff. 48 00:04:01,820 --> 00:04:06,310 How can we make use of a GP you or you in order to do this. 49 00:04:06,320 --> 00:04:11,150 You want to go to the runtime menu and select change runtime type 50 00:04:14,150 --> 00:04:17,170 as you can see there are two select boxes here. 51 00:04:17,300 --> 00:04:20,700 The first one lets you select which python version you want to use. 52 00:04:21,350 --> 00:04:25,190 So we'll be using Python 3 for this course. 53 00:04:25,260 --> 00:04:28,560 The second lets you select what kind of device you want to use. 54 00:04:28,590 --> 00:04:37,460 So that's either none which is the default or GP you or TB you now note that sometimes the GP you or 55 00:04:37,460 --> 00:04:40,850 TB you might not be available. 56 00:04:41,000 --> 00:04:47,360 This is because these are shared resources so your fellow peers taking this course and other machine 57 00:04:47,360 --> 00:04:52,700 learning students and researchers all around the world might be using Google whole app and we are all 58 00:04:52,700 --> 00:04:54,280 sharing these resources. 59 00:04:54,350 --> 00:05:00,420 So if our usage of these resources hits the limit of what's available then you might not have a GP you 60 00:05:00,420 --> 00:05:03,270 or you available when you need them. 61 00:05:03,620 --> 00:05:08,630 For this reason some of the code you'll see in this course may be done on my local machine as well. 62 00:05:08,720 --> 00:05:13,520 But remember Python code works the same anywhere so it does not make a difference 63 00:05:19,250 --> 00:05:19,820 next. 64 00:05:19,970 --> 00:05:25,560 You can see that there are two main types of cells that we can create in the notebook code and text. 65 00:05:25,640 --> 00:05:31,940 You can click on either of these to create a new cell of that type let's click on text since that's 66 00:05:31,940 --> 00:05:35,600 a little easier and it's not really something we are going to use very often. 67 00:05:35,600 --> 00:05:37,120 So let's just get it out of the way 68 00:05:40,480 --> 00:05:42,770 so I'm actually going to delete the very first cell 69 00:05:47,440 --> 00:05:47,730 all right. 70 00:05:47,730 --> 00:05:53,760 So as you can see when I click this it creates a new cell with what looks like a rich text editor. 71 00:05:53,760 --> 00:05:59,430 You'll notice that it's split into halves the left half is where you enter your text and the right half 72 00:05:59,430 --> 00:06:01,950 is a preview of what it will look like. 73 00:06:02,040 --> 00:06:05,170 So let's enter some text. 74 00:06:05,200 --> 00:06:19,450 This is my title now you can click the little T Big T icon which changes it into header text so you 75 00:06:19,450 --> 00:06:26,680 can see that it makes this a little bigger and bolder appropriate for a title next let's enter some 76 00:06:26,860 --> 00:06:29,180 regular text. 77 00:06:29,270 --> 00:06:34,830 This is regular text 78 00:06:38,750 --> 00:06:45,070 and note that there are also these arrow brackets so it looks like it's going to let us enter code snippets. 79 00:06:45,230 --> 00:06:48,440 So let's try that. 80 00:06:48,630 --> 00:06:54,970 And so as you can see it makes the text a mono space font which is appropriate for code. 81 00:06:55,210 --> 00:06:56,710 Now there are some other options here. 82 00:06:56,740 --> 00:07:01,010 So you can make a link you can add images you can indent. 83 00:07:01,090 --> 00:07:04,590 You can add a numbered or bullet list and so forth. 84 00:07:04,630 --> 00:07:07,150 So if you're interested play around with this. 85 00:07:07,150 --> 00:07:11,960 Otherwise we're not going to mention it again. 86 00:07:11,990 --> 00:07:13,460 Next we have the code cell. 87 00:07:13,490 --> 00:07:16,970 So let's create one of those. 88 00:07:17,110 --> 00:07:17,420 All right. 89 00:07:17,420 --> 00:07:20,860 And as mentioned we're not going to write any fancy code in this lecture. 90 00:07:20,900 --> 00:07:24,950 We just want to do something simple to make sure everything works as expected. 91 00:07:25,670 --> 00:07:28,630 So let's start by importing num pi and map plot lib 92 00:07:43,100 --> 00:07:44,120 all right beautiful. 93 00:07:44,150 --> 00:07:50,030 As I mentioned earlier these already come pre installed next let's create a new code cell and make a 94 00:07:50,030 --> 00:07:53,950 sine wave. 95 00:07:54,060 --> 00:08:00,530 So first we need to create some x values so let's make X go from zero to 10 pi with 1000 points in between 96 00:08:17,310 --> 00:08:18,660 next let's make Y. 97 00:08:18,690 --> 00:08:19,890 The sign of X 98 00:08:29,070 --> 00:08:36,410 next let's create a new cell and plot what we just created so that's just peels heat up plot x y 99 00:08:43,300 --> 00:08:43,530 now. 100 00:08:43,540 --> 00:08:48,640 Since this is a notebook there's no need to call people to you that show since the plot will just appear 101 00:08:48,640 --> 00:08:49,840 in the notebook itself. 102 00:08:52,230 --> 00:08:52,500 All right. 103 00:08:52,500 --> 00:08:53,370 Very cool. 104 00:08:53,370 --> 00:08:54,990 Works just like a regular notebook 105 00:08:58,020 --> 00:08:58,700 at this point. 106 00:08:58,710 --> 00:09:05,520 We've convinced ourselves that Google collab can do the usual things you'd expect from a group a notebook. 107 00:09:05,520 --> 00:09:10,770 Now as I mentioned earlier one thing that's very nice about COLA is that it already comes with a bunch 108 00:09:10,770 --> 00:09:13,380 of useful libraries pre installed. 109 00:09:13,380 --> 00:09:17,790 In my opinion this makes Google collab way better than Jupiter notebook. 110 00:09:17,820 --> 00:09:23,260 And if anyone ever asked me to write in a notebook environment I would choose color by default. 111 00:09:23,940 --> 00:09:26,910 I'm not a big fan of notebook but I am a big fan of collab 112 00:09:32,230 --> 00:09:38,110 so here we can see that I've written some code to try and import a bunch of libraries specifically these 113 00:09:38,110 --> 00:09:42,420 libraries are libraries that have been using my causes some more than others. 114 00:09:42,550 --> 00:09:46,120 Some are pretty rarely used so you might not expect that they would be included. 115 00:09:46,180 --> 00:09:52,720 Libraries like word cloud which we've only used once so far and yet if we look we see that everything 116 00:09:52,720 --> 00:10:02,050 I've tried to import here does not throw an error so this tells us that these libraries are indeed available. 117 00:10:02,050 --> 00:10:07,280 What's interesting to me is that some of these libraries are not machine learning related at all. 118 00:10:07,390 --> 00:10:12,760 Of course we've used them in my courses because they are generally useful as Python libraries but it's 119 00:10:12,760 --> 00:10:16,930 nice to see the folks at Google also make use of these same libraries. 120 00:10:16,930 --> 00:10:18,760 And so thought to include them. 121 00:10:19,420 --> 00:10:26,610 So here you can see the usual stuff such as psychic learn name pi PSI pi map plot lib and pandas. 122 00:10:26,620 --> 00:10:31,960 We also have torch and piano which is surprising because they are competing Deep Learning libraries 123 00:10:32,350 --> 00:10:36,520 and development for the ATO has been stopped for a while now. 124 00:10:36,680 --> 00:10:43,910 We also have seaborne word cloud beautiful soup which is for X amount CML passing requests which is 125 00:10:43,910 --> 00:10:52,160 for making a CTP cause network X which is for graph functionality CV 2 which is for open CV and gym 126 00:10:52,160 --> 00:10:54,170 which is open a gym. 127 00:10:54,170 --> 00:10:57,490 All in all very impressive and much more than I expected. 128 00:11:04,620 --> 00:11:09,410 So there's some final caveats to Colette that I want to mention first. 129 00:11:09,510 --> 00:11:12,860 The main thing you have to remember is that this is the cloud. 130 00:11:12,870 --> 00:11:14,780 So these are shared resources. 131 00:11:14,940 --> 00:11:21,060 So one way this affects you is if you leave your notebook alone for a long time it will become inactive 132 00:11:21,090 --> 00:11:26,790 and disconnect any computation that you may have run earlier won't be saved. 133 00:11:26,790 --> 00:11:33,230 So for example if you define a variable A equals five AND THEN YOU COME BACK LATER AFTER your notebook 134 00:11:33,240 --> 00:11:42,100 was disconnected and you tried to print a it'll say a is not defined so you see this notebook has disconnected. 135 00:11:42,340 --> 00:11:48,370 So let's say to reconnect and I print a it's going to say a is not defined 136 00:11:52,910 --> 00:11:55,680 another way this affects you is that you might run out of memory. 137 00:11:56,210 --> 00:12:01,520 So if that happens you might want to try running the code on your local machine instead. 138 00:12:01,640 --> 00:12:06,050 And as mentioned earlier the GP you A.P. you might be unavailable. 139 00:12:06,140 --> 00:12:12,920 So either you can run your code without the GP you or TB you or you can run the same code locally as 140 00:12:12,920 --> 00:12:19,280 always options you had previously are still available for example you can provision a GP you instance 141 00:12:19,520 --> 00:12:27,440 on a W.S. which if you choose the correct EMI or Amazon machine instance will come with the usual libraries 142 00:12:27,440 --> 00:12:28,460 pre installed also.