Original subtitles

1

Hello and welcome to the practical applications of this course I'm super excited to start with you and

2

I'm going to start with some good news right now and this Statoil we are going to install everything

3

so that at no time you have any installation issue.

4

We will install all the packages all the libraries anything that we will use to execute the codes and

5

execute the models.

6

So basically this is just the installation tutorial.

7

And then it will only be about coding and executing.

8

So this is a new feature we're very excited and proud to give to you.

9

I'm talking about the virtual platform this virtual platform is the combination of a platform that will

10

be hosted on the yes Web site.

11

It is right now in development and a virtual environment that will contain all these packages.

12

You will need to execute the codes in the models.

13

So right now and this Statoil what we're going to do is we are going to install an icon that because

14

we will connect this virtual platform environment to Anaconda.

15

And then in a flash light we will be ready to start and execute the models.

16

So let's do this let's start by downloading Anaconda and then we will install the virtual platform and

17

then we'll be good to go.

18

So right now I'm on Google and the first thing I want to do is to go to the Anaconda Web site.

19

There we go.

20

And I can digest that and I can then Google you go to the first link and I can get loud and then you

21

go to download and I can there right here.

22

We go.

23

And then you choose your system.

24

So when those macros are Linux I'm on Mac OS so I'm going to download Python 3.6 version.

25

You must absolutely download Python 3.6 because it will not work on Python 2.7.

26

So that's important.

27

And then once you're ready you click on this download button and the download will just start.

28

Right now it's just starting.

29

And one minute left.

30

Perfect.

31

We will have the ANAC on that platform in one minute.

32

So that's for a teacher to get a teacher.

33

But you can click on those things that's very important.

34

We won't be needing this.

35

So so far what can I tell you.

36

Well and I can There is a platform that contains several ideas.

37

We will be using one of these ideas which will be spider because it's like a studio you have a viable

38

window where you get to see all the variables you create your file explorer.

39

It contains some interactive and dynamic eye bison console so very efficient to execute the code.

40

This is definitely the one we want.

41

Everything will work like a wonder.

42

So here we go download is done.

43

So what I'm going to do now I'm going to go to my download folder where I can find a neck on the package

44

folder that I just downloaded.

45

So let's double click on it then we will get the installation steps so it can continue continue continue

46

agree and still for me only That's fine.

47

Continue and install arise and right now it's installing it will just take a couple of seconds and it

48

should be ready very soon.

49

All right.

50

Almost done writing package receipts is perfect.

51

It's just installing and I can there with all the pages are going to do that.

52

But we will be installing more packages.

53

Thanks to the virtual platform.

54

Here we go.

55

That's done.

56

Now we're going to click on close and you can choose to keep it.

57

That's fine.

58

Just to keep the download file.

59

OK.

60

So now Anaconda is installed but we're not going to open it yet because we'll be installing the virtual

61

platform.

62

So now stay with me and let's do this.

63

It's going to be very efficient.

64

We will just executes one line of code in the terminal.

65

Or for those of you who are on Windows in the second prompt.

66

So first of all let's go to the computer vision is that folder that you downloaded in the previous To

67

toil and let's go inside and let's go inside the installations subfolder this installation subfolder

68

contains three way e-mail found one for Linux one for Mac and one for Windows.

69

You don't have to do anything with those.

70

Why not file you can open them if you want but please don't nullify anything.

71

These are just why a mail file that contains all the dependencies packages libraries that are going

72

to be needed for the morals that will be executed in the course but you don't have to open them.

73

What's We're going to do now is open a terminal or the Anaconda prompt.

74

So for those of you who are on Mac or Linux you just need to open a terminal.

75

I am on Mac so I just type command space and then terminal here.

76

Answer.

77

That's the terminal I suppose know how to open a terminal on Linux you can find it very easily at the

78

left of your screen.

79

And now for those of you who are on Windows what you need to do is to go to the lower left corner of

80

your screen click on the windows button then find in the list of programs Anaconda's 3.

81

You go inside and then you will find an icon that prompt you open the prompt and then you will get a

82

black rectangle like this a black screen.

83

And now we are on the same page.

84

We just need to type something that will install everything and then we'll be good to go.

85

OK so now you guys on Mac and Linux have your terminal opened and you guys on Windows have your and

86

can that prompt open.

87

And then we're going to type a command line to install everything.

88

But first we need to be in the folder that contains the Y MFL.

89

So we have two choices now.

90

Either you find this folder which is the installation folder through the terminal you know by typing

91

CD then desktop because the computer vision it is that folder is on my desktop and then computer vision

92

it is that then instillations that we go and then you press enter but maybe that's not the best way.

93

If your computer vision is full there is somewhere hidden in your system.

94

So in that case what you can do also is just take the way a male that corresponds to your system.

95

So for example my way email is Mac so I'm copying this I'm putting it in my main folder which is an

96

law here I can paste that here.

97

All right.

98

And now since my terminal is in the right folder along which contains my way e-mail Well I'm ready to

99

type the command to execute the code.

100

So as you want it's whether you are comfortable not with navigating inside your system that's the two

101

possibilities.

102

The only thing that matters right now is that your terminal or Anaconda prompt is connected to a folder.

103

And the way a mail file corresponding to your system must be in that folder.

104

So if that's the case what you only need to do is type Konda

105

create then minus f then virtual platform.

106

And then if you are a Mac you type Mac.

107

If you are on Windows you type windows and if you are on Linux you type Linux.

108

I am on Mac so I have Mac and then don't forget the dot.

109

Why.

110

Well that's all you press enter and this will install everything.

111

And at no time will you ever have to install anything else.

112

So let's press enter and let's see the magic.

113

There we go using an icon that API fetching package.

114

Metzer that's out and it's going to install everything now and this shouldn't take a while.

115

This should take like one minute to two.

116

Let's see something back package specifications and there we go.

117

Isn't it beautiful.

118

Everything is being installed.

119

And we won't have to bother about this after that.

120

So that's pre-reg saying that's a new feature.

121

That's the virtual platform behind the scene.

122

You know we're installing a virtual environment.

123

But of course then you will have the virtual platform the graphic virtual platform on the LDS web site

124

which is still in development.

125

So beautiful it's going to take maybe one more minute.

126

So let's see.

127

So as you can see it's installing some packages like f f MPEG because we'll be working on some videos.

128

It's installing pipe and 3.5 3.6 because by then 3.6 have some issues with by.

129

It's installing by George of course because it is the most powerful to implement neural networks.

130

It's installing open C.V it's installing side by it's installing.

131

Well everything we need and even some more stuff that like you know very useful bits that we won't be

132

using it but it's because of the dependencies.

133

You know when you and someone package you have to install some different packages.

134

So but anything you'll have everything you need.

135

OK.

136

So I think that was the last one to install.

137

And now you just need to wait a few more seconds and we will be ready to start very soon.

138

I'm so excited and so excited because then I will show you right now the folder that contains all the

139

models and we will test the virtual platform to make sure that it works we will test it with the first

140

model.

141

So let's see let's see what's going to happen.

142

All right.

143

Yet that looks like it's the end.

144

It's going to be the end in a matter of seconds.

145

And there we go.

146

Yes.

147

Perfect.

148

So no error.

149

Everything is fine.

150

It tells you some stuff like to activate this environment use source activate virtual platform.

151

Don't worry about that you won't have anything to do about that.

152

I will show you what to do right now.

153

It's so easy.

154

So let's do this.

155

You can close your terminal or Anaconda prompt now.

156

And now what we're going to do is open Anaconda.

157

So for those of you who are on Windows you can find it by clicking on the windows button at the lower

158

left corner and then find an icon that 3 but you have to find an eye can the Navigator.

159

And here on Mac I'm going to type Anaconda navigator like this.

160

It's in my apps.

161

And there we go.

162

And I can that is about open initialising.

163

And here is Anaconda and now the only thing that we have to do to connect to the virtual platform.

164

There is nothing more easy.

165

You get two applications on then you click on this little arrow.

166

And here we go.

167

Here is the virtual platform you click on the virtual platform.

168

It's loading.

169

And now we are connected to the virtual platform inside Anaconda.

170

That's beautiful.

171

So now the only thing that you have to do left is just click on launch you know that's the idea we're

172

using spider.

173

You click on knowledge and you'll be inside of a beautiful studio which will use to execute the models.

174

There we go.

175

So I'm going to quickly present Spider.

176

This is the area where you have your code.

177

This is the bison console.

178

Make sure that you have the item console and not the button console.

179

Here you have three windows Falak 4 or you have all your files your folders.

180

So we will find the computer vision areas that folder will explore where you have all the variables

181

you create and help if you need some help sometimes you will get the help to get some informations of

182

the object's classes we're going to use.

183

But the first thing we're going to do is we're going to go to explore and find the computer vision it

184

is that folder.

185

So as you saw it is in my desktop there it is computer vision is that.

186

And now we're going to test the virtual platform to make sure that it works.

187

So I'm going to click on computer vision it is.

188

Let's test for example module one face recognition.

189

So in each of the folder modules you will have to code the commented version of the code and the code

190

without comment.

191

So the command version of the code is the name of the module face recognition and this quirk invented

192

and the no commented version of the code is face recognition and is going to comment.

193

You can find the same for the other module to object detection object detection commented object detection

194

of Cummins and for Gannes DC again come into DC again no comment.

195

This again means deep convolutional again.

196

And then you have some other foules which are some other bytes and files or XML files or some folders

197

that contain some tools that are required for the models.

198

So for example here for face recognition the tools are two Cascades one cascade for the eye and one

199

cascade for the face.

200

These are ex-MIL files and theres just to be able to detect the eyes and the face with open city but

201

the code the heart of the code is one of these vowels.

202

So you have to come into the version of the code and we together are going to retype the code line by

203

line so that you can understand what's going on.

204

But for example I did a double clicking on face recognition commented and we're going to test it out

205

right now.

206

So as you can see each line of code is commented that's the same for the other code and to test it we

207

simply need to select the whole code by pressing command or control a and then to execute.

208

We simply press command or control enter and there we go.

209

Is it working.

210

Yes it is I'm going to show that to you right now.

211

Here it is.

212

That's the result of Module 1.

213

So hi guys welcome.

214

Welcome to my place.

215

This is actually where I'm recording all the tutorials.

216

So this is face recognition as you can see we detect very well my face in blue and my eyes in green.

217

So that's one of the models we implement and of course that's a very classic example.

218

But don't worry the more you will progress and of course the more you will get some very very powerful

219

models especially the one for object detection and the one from module 3.

220

And also this virtual platform will be updated through time we'll add some new stuff some new models.

221

So make sure to keep in touch with it.

222

All right so I'm going to press on cute to exit this.

223

There we go so that works very well.

224

We don't have to test it for the others.

225

I actually tested it 10 times to make sure that it works.

226

I tested it on Linux Windows and Mac.

227

So you should not have any problem.

228

All right.

229

So I hope it's clear.

230

I hope you manage to get to where I am right now.

231

Is it getting this code if it's not the case.

232

You can ask some questions in the Q&A.

233

We will assist you.

234

There is a great support team that will be very happy to help you.

235

All right so I can't wait to start coding all this.

236

And until then enjoy computer vision.