 Hedy technical stuff you've been going through the morning and what's going to come I'm not a programmer by training. I wrote my first program about five years back So and you know back when I study There were no programming courses for mechanical engineers Then about five years back First first program that I wrote was in Matlab. My wife taught me how to write a function That's how I started out and I Coordinate this area of machine learning. I work for a company which We have a division which makes Products for the investment trading domain. So we make the products using machine learning So I got into this area and for a year. I used Matlab I found that you know as a new as a new VNU programming I found in Matlab Extremely easy to learn extremely easy to work on So it's after a year that my colleague introduced me to Julia Then I started working on Julius Victor over to Julia from Matlab. That's that's like about three years more than three years back Yeah, and I have not used anything else after that So we're working on Julia for the last three years and that is really the first programming language that I worked on So probably that's what we all wanted to highlight. Yeah So I guess to start with in Matlab one So when I speak to Matlab to Julia One issue which I faced at that point was that it was not enough information on the web On Julia as there was Matlab for someone who's starting out. Obviously, that's the only way to learn Now I am in over the period of time I've seen that the information the rep has dramatically increased now I don't see an issue of not getting not finding or you know having to ask someone for You know how to how to write a function or how to what function will work where? The other hand what I found really good was that Matlab When I used to write functions It was kind of slowing down and scripting was faster I do not know whether I was doing something wrong. I don't know whether it's the same right now But at that time scripting was faster than when I have lots a lot of functions Which is not what you know, which is we didn't really pass it. I wanted to write function So that was one thing and when I spit over the Julia realized that you know here It doesn't doesn't work that functions are better. You write functions and that's better So that that was so that was a band-aid is a band-aid I saw when I switched over three years three and a half years back and So I've worked exclusively in Julia as I mentioned and I've used a couple of machine learning packages There are quite a few packages for machine learning and I really can't evaluate the pros and cons of these this package versus that of others What I've seen what I want to point out Is that the packages in Julia you have code that is transferred Which is not the case with some of the languages that I saw Matlab is a case in point You know Matlab has a lot of packages and obviously more than that what Julia has right now But some of the packages when I want to see the code Behind I wasn't able to that kind of impeded purpose So in machine learning what I've seen is that you know mission learning the set of tools right many of you might know and These tools can be used by people who you know want to learn machine learning and who knows the maths and You can learn it But the key thing I believe is to apply the tools Innovated me in a certain way where it will fit into your domain whatever application you want to apply And for that you necessarily have to make changes in the The code as an in the package code when you use package when you use a package You have to make changes and you have to There'll be a number of times when you want to make changes to the package code. I want to quickly Mention a couple of examples of my experience of doing this these are little packages, which I have used and It's a it's a if you look at the code for these packages, they are crisp. They are short. They're not elaborate They have enough functionality. I mean, we still they're still being built and there are times when these change people contributing so for example decision tree, I mean I I I hope people are familiar with Basic machine learning so When you use a random forest within decision, you have a package in Julia, which allows you to do random forest Most of the time the tree is a train. You train the trees And then you the trees are generally trained based on a certain of people function And you need flexibility for the activity function You don't want to have a single and usually Any standard package any standard training mechanism will create it based on accuracy How many instances are classified character? That's that's the usual objective function used to training forest Or or a tree But you might want to do something else because when you use it machine learning when you use Maybe I should give an example. Let's say you want to you want to you're facing a problem of predicting How probably the person is person is to have a heart problem So you have a lot of features and you want to build a decision tree and decision tree It kind of helps you to See whether he's a he's in a high risk or a low risk a typical machine learning problem And if you have a lot of features would be like the cholesterol level or the fitness regimen When you try it you might not really want to stick to The accuracy for duty function alone might want to bring something else some other and there you need to do a lot of changes and These lot of changes become very few changes when you go to these tactics That's the biggest advantage of scenes. You start to write a couple of lines And then write a few lines to change the And there's validation validation is like when you train you want to see how well it is performed I'm writing this standard. There's standard millions of values and full You want to change it you might want to change that might want to wait You don't want to wait everything all the instances similar way. I want to wait certain instances in there With a certain measure Like for instance in finance I work in finance You don't want the accuracy as in the number of instances that are classified correctly I might want to base a model on returns So that you have a global model. You don't have a local model So so if you want to base it on returns, you'll have to change the duty function So that's that's the biggest value that I have seen in my experience of using Julia where you can just change code in packages easily, which is helpful and That's probably what machine learning I don't want to get any further. That is the other Package you can use text analysis, which is basically sentiment analysis using textual features You look at Documents look at all the words and try to extract information. We can help you in decision making. That's text analysis So that's another package and we've used extensives. The same thing goes for that package also If you put if anyone is interested you would look at these packages and see how How Chris the code in there code is in there the second part of it, which I want to talk about is about data manipulation, so As when you work in the finance domain more time goes in data Action model that's what usually happens because you deal with large arrays of data and I mean what number of Languages, but what I've noticed what I've heard from my colleagues with Sarah Julia is that it's much easier to manipulate arrays and manipulate data to make it in the way you want So that's the other advantage. I could just quickly Show you something this is a very simple thing, but I just want To show it just for you to understand how to deal with large arrays What is what is important here is you have to do a lot of input output as in reading from the from the file on your heart is and writing in the file So that is very easy in here I mean you can do it from any program in London, but it's the syntax is fairly simple in there. So I have So I have a file saved here demo to jail and He's picking up a part that particular part read that file So since I am using dollar part because part is already You have 24,000 25,000 rows and four columns. Yeah Now let's say I want you know get out get it off the labels with the column labels. So This is You see I'm getting rid of the first one first So now I have Without the hand Now, let's say I won't let's say the fourth column fourth column has probabilities I want to pick out all the rules which have a property greater than 0.7 So I get so I get all the rules which have Probabilities more than 0.7. So which is around 9,000 rows Let's say I want to make the first column in the years So I have Now I have these as individuals Let's say I want to Convert these dates I mean, this is a very important thing because in finance we deal with lots of dates all times use so date conversion becomes a Major work sometimes because you get dates in various formats So here I'll just use the package So you get a package Three date so I just have to put the The format the date format here, which is like this y MD For this So that's done Now I have dates in the date format instead of the string that we had earlier Now I can kind of use it like I can use it to you know find the kind of number base between two rows Data Let's say 10 comma 3 which is the 10th row the rate column minus data first first row So I get 28 days. So I can basically now do anything with this dates Now, you know if I want to do something like if I want to concatenate to It's it's it's very simple if I want to write it up to a file It's just right CSV data that's done and So the reading and writing which Reading and writing is this is very simple which is which is adding the best part for people like us who work on a lot of data Yeah, and that's probably what I want to say Even if you want to see something else with one here and try All right, any other questions on using you there. I probably don't need a bike So I just started with machine learning and so far I've just built static models Well, there's a data set and I train it and I get it. How do I make something dynamic and is Julia have some packages that make it dynamic as in you dinner comes in and he trains and whatever So the most is the model you want to water me Not all of it's more like they are dynamic automate I can write a batch file or something But is there something like that in Julia already? Yeah So I mean what we do is so once you have the model ready once you have the code Yeah, you put the new design. That's what we do. Okay. What you're saying is To make it dynamic anyways before to give new data So the problem is this thing if my data is 100 GB and every day I get one GB So if I'm running the whole thing, then I'm running basically appending this data and then running the whole script again and again Right. Why can't I just run it on like the new data is appended and this Some sort of connection only on that No, is there something like this I'm starting very And so first of all you train the model with Another instance and you have a new instance And if you're going to drop off the first one, I Don't think it's going to make a huge difference I mean usually Unless it's real it's real if it's an outline, you don't want it to make a difference Because you want the model not to fit to that data, right? The idea is not to over So usually when you add one single data point, it's not going to I mean generally don't do that. They generally kind of they say change the data If that's our hundred twenty twenty percent has come new we add that in training So how does your model work in the R? Can we take the software and so fingers this process is called online learning separate package That's not a Julia question I Julia Get a streams the data that is coming in You can there is I mean you have a database where it's Online streaming Get out of your stream frosting Support for speed There are various streaming packages actually In pretty much basic Julia code You can just stream things through We have sort of making too much of a fuss about it because the performance is good so you don't need an elaborate system But there are there are online statistics The streaming statistics packages that people have written which make it a little more convenient Yeah I Think you're short of time