Hello, my name is Manu Mitra and I will be the instructor for Microsoft Azure Machine Learning Studio. Okay, so we will be more are developing algorithms. I mean oh really not be developing. So I'll be giving an example. So take it as a tutorial. Although I can do multiple examples we'll go like three to four examples about how do we do what do we exactly do in this machine learning.

This Microsoft Azure Machine Learning Studio gives you a very good idea on a new algorithms without actually going into the code. So I'm not touching any code. There is lot of code or there is the C c++ they are okay, this is not of course, you can Use, but in this specific classes, I'll not be using any of those codes because this will take time because this is running no this is actually I will show you the algorithm we will discuss one algorithm and we will see how this works. Okay. Oh actually, you know, to be honest, you will get an idea how this may machine learning works, you know, how not machine learning versus like, how you can develop an algorithm in like, if you want to build up in machine learning something, you have to write in an item, right? You have to come up with an algorithm.

So, in this class, you will be learning how to actually get an alligator to get how to how to get another letter based on your requirement. Actually, it's a completely visual based on your requirement. How do you actually want your algorithm to be and how You are from how do you want to look like know how to you know, how the system should look like and how you should. So, we predict from the data, so, we have six of data and we predict from that and develop an algorithm to see that how this machine learning will help you in the long run, okay. So, this is free, I mean this, this will this will be free. I will be giving you access for this.

I mean, this site is free engineering. Okay, so, first thing first on the problem is this is from Microsoft. No, no, that's not the problem. Okay, maybe it's not the Microsoft is not the problem. The only problem is with the email id. You cannot have this any Gmail or hotmail or yahoo email ID for signing up.

Probably you might know you don't want to do this. So What you want to do here is you want to get an email id if you don't have money if you have email ID from my Microsoft customer name good or if you don't have what I recommend is go to outlook.com probably a nightmare come here Okay, oops, looks like I've already signed it so no guess what I'm signing out to show you actually do you know first thing is if you don't have Microsoft email id then you want one right probably the Indonesia the new scope or something is wrong. So the few that's correct Okay. Now go to Outlook that okay, they haven't different okay outlook libre calm probably I will include this in this material where you can No, I'm not going to show you how to create me me lighting. Okay, I like to show that but the point is you can create an angle it like this.

Okay? I will include this what I'll do right now, I'm actually no include this is this. So I'll be giving you access for you to get an access. Let's see email id registration. So I'm not going through, you know, this registration of email id of Microsoft but you have to register it to get, you know, to do use this Machine Learning Studio. So we see like, I not be developing lot of algorithms, a lot of things I'll be going to this tutorial and I guess I also have this with this PDF to send me a PDF Which you can go through it.

I feel wanted to give two tutorials to be honest. But I have the data to see this data, we could upload this data and predict the values. But unfortunately that that string too much of time that it's almost like more than 510 minutes. So we don't want to waste time on that, sir. All right. So, you know, created go ahead and create an Microsoft email id if you don't have one.

If you have one, you don't have to. You can use the similar one out, it's up to you. How do you want to do it? But once you create an email id, you know, I'm not going to do this. You know, you're not going to do it because I already have it. You have two options, hotmail or Outlook or outlook, I probably go to Outlook and create an email id.

So you can create an email ID from a Microsoft once you have done that that departed from That's the tricky part. If you don't have one, you have to create one. Okay, so now let's, let's assume that you have already created or you already have one, that's fine. Now we'll go to here studio dot Azure ML dotnet. Okay, this is good. I'm to do a short demo that this is Microsoft Azure Machine Learning should you here what you do what exactly you do is you create an algorithm here, the whole point of the class is to create an algorithm.

But I know we only to be honest, I'll be really showing you one, two, the max three and we have the some of those already inbuilt algorithm. We want to let the reason that not that I don't want to it's like, we'll keep it short right with you. Don't worry, you know, the point is to Understand how to create an avatar and then you know going through so that's the point you know we don't want too much clutter like 1020 classes. But the point is you have to use your creativeness. So here the creativity is very important so what I'll go ahead and we have this option so you can go through those videos and you know, let's let's sign up but we'll see here. And, you know, we will login once you log into that you have to use this Microsoft email id once you log in, so since I already logged in, so I never asked me my email id and password but you have to but you have to log into your email id and password.

Okay, good. So we are into this system. Okay, so here look at the system let's let me walk you through all this options here. They're not easy options this thing is to working on what is it working on? I'll go ahead and delete this one for now. But okay, good.

Um, so this is my experiment I've done quite a few experiments and experiments you have samples to cables from books or let's not overcome this first. So um, you know, we'll go to First thing first we go to administration a little bit of administration work here, okay. It administration is little bits, small here, okay. Probably you might not owe us this as for corporate environment, but you can use this you know, as Personal personal experience, okay, personally, it's free. But if you are using is the carpet you want to use in a company or something, probably you have to buy this enterprise license, otherwise it's free. This is the Shenzhen good concept on how to develop an algorithm.

That's the whole concept here. Okay. But Hi, who takes some time to show you what are the various steps you can perform, we go to one example, actually to a young boy and two of those involved and then we wrap it up. So probably fiower class will be added, but at the end of the day, you will be getting some idea on how to build when Garrido based, it's about prediction how, what you want to do, and what algorithms you can use this just in tutorial kind of thing. But it gives some some information. So all right, so let's go to administration part we'll talk This this they notice this left hand left sidebar, we have this various options have not forgotten it.

We go to this Okay, let's go to say no Guess what? Here you have this option if you go there, I cannot upload my picture for some reason, they're there, they have no option actually. So, you have this option on this is fine and this is feedback option. Maybe you can provide feedback. I will be going to this documentation to understand this So, I went to the document and show you this whole is covering the PDF document which I will be setting up. I named this as let's take as a Udemy workspace and we'll update it and lifan professional no we could build any workspace and this workspace I is here and we are in free mode.

So we are free, so you're good. And don't forget to save this option is here, right here. Click on Save, and it got saved. So this is supposed to change to Udemy. So it didn't change. We just refresh it.

So hopefully that should be good. I am hoping that that's good. So it takes you to a Udemy workspace. Okay. This is a I mean, this this naming convention, actually. So you have this option of yoga is for everything.

That's the same thing. Okay. So here, this is administration point in the settings, we are actually in settings. I'm just talking about this administration, probably. I will, you know, probably be the mobile module to this button in the next class. Probably I'll explain some of these in detail but we have this idea.

Of authorization tokens, go ahead and click on it. So this is an authorization token and can be generated if you want secondary authorization token. So what exactly use of this integration. So here, whenever you want to use as a third party and you want to give this code to access this, but you're not using any third party core for now, right? This as far as I remember, this was used when I was used to I used to have this website, right. So this has spoken it, or whenever I used to use this third party thing like like web services, like I want to call some function, not exactly function that some particular let's say it's a capture recapture that then equals capture on the chin to recapture.

So you want to view this thrashing code that was generated from my website. So that is the reason you have this one. Author That you're giving me an authorization to access that application into your web page. So that was the use of this. Okay, so that this third party integration normally for third party integration, okay? To make it easy if you're using some third party application tool, so you can use that one.

So let's go to where we've we have done with the names, right, so names is good. And our iteration the users are curious. This is a me I mean, this is the main thing it's like, you know, this is where you can invite users let's try to invite them a few. You can invite probably they want to have this as Microsoft account. Otherwise, probably they might have to create Microsoft see this Microsoft in a design, they're owning a lighting They didn't, they weren't like someone@example.com but they say that you have to sign up for Microsoft email id otherwise they don't let you access so so you have to have this Microsoft email id and then you specify it here and you can view access. So here's the thing.

They don't have much convoluted access. Fortunately, this is a new thing. So they don't have much only users and owner. So owner is like a super admin they can delete with access, of course many user can do, they can delete and they can edit and all those things user have limited access and they cannot delete and all delete the you know, whatever. You have created whatever algorithms you have created that so so this is you know that's pretty much about it me natural technical in to be honest. This is that say It means you can invite so I mean, this is pretty simple for you as Asian, either be owner or user.

So if you're working as a, I don't think I really have problems if you're working as an AP, let's say you people are let's say your friends want to work and you know, you want to work on that. So, probably you can invite them and you know, so, here is this is an option not new. This thing is new for every project, you shouldn't this new thing comes up. So, this is the adoption This is the option here you invite it and then you can remove or so you as a super admin You can also remove if you if you are an admin. So whenever you sign up with this, that means you are super admin, that means you you have the access to add user and remove user. Okay, let's so I signed up with I have to be access with this way actually, I can't For practice so you, you can have this one email ID for your super admin access and your friends can you know you can invite your friends and probably they don't charge should be free.

But if they're charging it make sure don't do it. I have never tried that probably didn't do that. So you have data gateways, okay, you can add a new gateway. That's good because this is not this is for free so you can add it. Okay, so we don't want that either. So that's pretty much about administration work.

Okay, you can delete it that workspace this not deletes This not only deletes your workspace but the whole whole machine learning thing. Whatever I got them co created swag utility. Again, you have this 10 g be soft space, so that whatever algorithms you want to create it, you can create an action. So you know, so what exactly is this idea? Here in this Microsoft Azure Machine Learning Studio specifically, it's basically a flow chart. So if you have done some work in ETL ETL work to run a ETL developer or something and if you are using Informatica, this will be a piece of cake.

But otherwise, you won't have to know it. We'll go to one example. And I will show you some of these options it will be like ETL process, okay. So, so that's that becomes an algorithm actually. So So what is an algorithm it's kind of programming, but here we are not using exactly we are not touching any code here and this any of my classes of this specific class. We are not touching any cool.

For now, probably, let's see if we have probably we can look at rate, but our code probably does support a lot of good things. distributors do support our core, Python and the cc c++ I get. So a lot of you can use a lot of code in this too, but you're not doing that actually. So that's pretty much about this administration point of view. Okay, so, um, it's actually already 30 minutes. So I guess I will stop here, because I wished This was further because we have a little bit more on this.

Okay, and we and actually one hour actually takes like one GB of data array to one two GB so It gets limited here. So that's the reason I keep sharp. So, probably we stop here. And that means the next class will go through this options. Okay? This was like kind of administration, you know, you go through this, you change this workspace and hit save, we'll go delete it, right?

Yeah, this will permanently delete this workspace and whatever content you have created for the gardens and all those things who know that gets deleted. So, um, so we stop here at this point. And we know we'll go through this all these options, right? We go that we don't only about settings, settings is actually here is kind of administration where we can add people in users use module you can i here. Okay, so this is pretty much about administration. So that's pretty much single shot.

But let me give you some, some background at you Why do you want to use this machine learning? simple equation, right? Why do you want to use machine learning? So the thing is, you want to use an algorithm. So this kind of things, right? Let's take a specific example here.

Okay. Now there is a lot more things you can do with this machine learning probably you can use a drone. Let me give a specific example drone to move you know from here and there little bit. Bit, the no user interaction with nothing and drone can do all by itself. So that's one of the example way you can use this on machine learning. But in this scenario, we will not be going to that level of specific but here we are dealing with the data.

So of course it goes based on data. So we will be working on the data and more prediction, how do you predict the data and view the chart and see that error ratio and it's in? Or is this matching for you? It's not matching. And we'll look into it and see that, you know, how does it work? So that's one of the example as I told you, for machine learning is, I mean, it's to at the end of the day, it's kind of corny, right, it's kind of point.

But in this specific scenario, we'll be predicting data. And based on that data, we know you could, it's actually at the end of day to work you want, what you want, and how could we use that particular algorithm, a particular mathematical operations to perform your work, right at the end of the day. That's the whole point. So, so that's pretty much for the administration. So I put it as this as an administration. Okay.

So thanks for. Thanks for watching, and I hope you liked this video and see you in the next class. Thanks