How to Get Started with Machine Learning & AI

hello everyone and welcome back so in this video what I'm gonna be doing is telling you guys how to get started with machine learning so I see these questions all the time from usually beginner or intermediate programmers asking me where they should start and what they should do to get into machine learning and artificial intelligence so in this video what I'm gonna be doing is telling you first of all what language you should be using slash learning what kind of Mac background you need to do this and then what process you should be following in terms of what should you learn first and where should you kind of move from there to get better and better with machine learning at a beginner level so with that being said let's go ahead and get into the video and talk about how to get started with machine learning before we get started I need to talk about the sponsor of today's video which is kite now kite is a free plug-in for your IDE or text editor uses machine learning to give you the best possible completions for your Python code it's capable of completing entire lines entire function calls and it ranks all of its completions by relevance so you get shown the best ones first even as another feature called intelligence snippets which allows you to quickly tab to the different options and choose which completion you'd like one of the coolest features that comes with kite is called co-pilot now what code pilot does this provide one-click documentation it shows you information about modules classes methods and functions based on your cursor location now the best part of kite is that it's free and you can download it at the link below alright so the first and probably the most important thing to talk about here is the language that you should be using for machine learning as a beginner just getting started now there's tons of languages that are designed specifically for machine learning like our than some other languages that I can't think of a top of my head and you can even do it in things like JavaScript and Java and all these other languages you can do machine learning and artificial intelligence in now the language I'm gonna recommend is Python now the reason I say this is because it's kind of the industry standard for machine learning and artificial intelligence and if you get a job in this field chances are you're probably gonna be using Python to do most of your work now it's great because you don't only use it for machine learning it's capable of doing many other things and it's probably one of the easier languages to pick up and start using as a beginner so that's why I recommend Python it also has the most kind of modules and support for machine learning and artificial intelligence and if you're looking for tutorials and trying to learn to do this you're gonna find the most support and resources in the Python language so I recommend starting with Python as you get more advanced you can move into more specific languages but python is a great one to get started with alright so the next thing to talk about is math now I know this scares a lot of people away because they always say you know I'm not very good at math can I get started with machine learning I know machine learning is a lot of math do I need to know the math to do it now the answer this question is no you do not need to be good at math to do machine learning but if you are it will help you tremendously now let me explain this you can go through and you can do any kind of machine learning tutorial you can even you know probably get a job without understanding how a lot of the math for machine learning works and you can use it fine and you can make predictions you can create applications you can do cool things but if you understand how the math works it makes everything so much easier and so much clearer is at the end of the day machine learning is not just this black box that you know something comes in something comes out it's a very complicated mathematical algorithm written by you know professors and math experts and all you do is use this map that's already been created for you to get the outcome that you want so if you can understand why this works and why things are happening the way they are you can be much better and you can have a much clearer understanding of all of these things all right so now that we've picked a language and we understand how much math we need to know for machine learning or maybe how much math we don't know for machine learning I'm just gonna kind of walk you through the steps of machine learning now the reason I'm doing this is because a lot of people that are getting into this don't understand what's actually involved in doing AI machine learning they see some cool video of some guy who made some awesome machine learning project did some cool artificial intelligence robot whatever it is and they want to get started but let me just walk you through what actually goes into doing this because you might realize quickly that maybe this isn't exactly what you want to do or it's a little bit different than you were thinking now the first step in any machine learning kind of project is analyzing the problem or what you want to do figuring out what is my machine learning model I'm gonna do for me isn't gonna predict something is it gonna create something is it gonna recommend information what is it gonna do understanding the problem is important and that's how you're going to pick what algorithm you're going to use now you do that it's time for data collection what data are we gonna use for this machine learning algorithm we know with machine learning artificial intelligence we need a ton of data ton like sometimes you could be using terabytes of data to train a model so what data do we need and where are we gonna get it from after we do that we have to clean the data and all of our data comes in nice and crisp like we might see in some data sets we need to clean this out we need to pick what we actually want and that brings us into the next step which is data selection we have tons of data but it's all of it relevant to our problem is all of it relevant to what we want to do answer is usually no so that involves now you know we collect the data we pick what we want and we have to pick the important things so we're actually going to use in our model now after we do that what do we do we have to pick what model we're gonna be using what algorithm we're gonna use how we're actually gonna solve this problem using machine learning or using artificial intelligence now after that we've picked what model we're going to use we need to train that model so feed all that data in and after that we need to validate the model we need to test its accuracy we need to see if this actually made sense and if it did something and then what we need to do is keep doing this again we need to keep tweaking the model and making it better and better and better until it can't get any better anymore so notice that in this the actual machine learning part of this like running the algorithm and training and testing the model is not a major part of this a major part of this is picking what algorithm we're gonna be using and picking the correct data and that's a massive part of machine learning and that's what's gonna bring me in to my next point here on where we need to get started with machine learning the first thing that I always say to people and the first kind of thing you need to do is make sure you understand the basics of the language that you're using and maybe even more than the basics kind of like the intermediate concepts involved with that language like objects and classes and all of that so for this video I've recommended Python for machine learning so if you're gonna use that make sure you know how Python works you know the syntax and you know all of that because these beginning steps involve understanding Python to pick out your data to clean it to select it to do everything you need to do because data is the most important part of our machine applications usually other than the model now once you've done that we've understood the basics and you feel confident in the language it's time actually move into machine learning now there's so many different areas of machine learning and artificial intelligence so I'm gonna just specify where I recommend beginners kind of start because there's so many places you could go now I always say to learn the basic algorithms first now this is not fun most people don't you know brag about using these algorithms but they're really important to understand and they are fundamental and if you actually want to understand how things work and kind of slowly progress in the correct order this is where I would say you should start the first algorithm I always recommend you learn is called linear regression it's a very simple anyone above the age of 14 or 15 should be understand how this works just based on some simple math principles and it is powerful in being able to predict some specific information now just before I go too far I do have a tutorial series on my channel that kind of covers the sequence that I'm talking about here if you'd like to check that out there is a card here but continue watching the video before you go over to that so the next thing that I say you should learn after linear regression is K nearest neighbors now some more advanced algorithm and is very very powerful K nearest neighbors is used for tons of different applications and even some of the most advanced AI we have uses K nearest neighbors to do specific things and to cluster information now this one's nice because it's kind of a different flavor from linear regression it's not predicting a number but it's actually classifying information into groups now you'll see how that works when you kind of learn this but I just want to tell you that these are different algorithms they do very different things now the next thing I recommend you learn is support vector machines now these are more advanced they do with high dimensional data and this is where your math is going to be tested a little bit and understanding exactly how these work now once you've done these three algorithms and I mean you've done them you understand exactly how they work you practice them on a few different data sets I will kind of allow you to move into whatever area you want now and I'm gonna give you some suggestions but what I mean is I always want people to get started with B's because this teaches you the basics of machine learning which I've talked about before data selection data cleaning the importance of data and really just you know having an appreciation for how difficult it is to make an accurate model so once you've done this now you can kind of move into the area that you're most interested in now I usually recommend starting with something simpler like q-learning which is actually a very powerful algorithm that's used usually to train things to play like video games it's known as reinforcement learning and that's its whole own strain of artificial intelligence and machine learning but once you've kind of done cue learning you can do things like the neural networks you can do like computer vision you can do all of these different things but start with those three things so linear regression K nearest neighbors and support vector machines and then move to where you want to go cue learning is good if you're trying to train video games to play things neural networks are just PI used for everything and there's tons of tutorials out there I have a tutorial on my channel or how to use neural networks as well and then there's some other different strains of AI that you'll start to learn as you move more into this field alright so for those of you that watch the video up until this point I want to start by saying thank you and I want to leave you guys with something to reward you for really actually paying attention and doing this now I'm assuming most of you have some experience with programming but if you don't you're brand new to Python and you really just want to get started I do have a premium programming course that goes through the basics of Python and covers everything that you need to know to get ready for machine learning if you'd like to enroll in that I'm going to be giving you guys 30% off just for watching the rest of this video you guys can use the code it's gonna be on Pierce somewhere and enroll in the description down below now with that being said I hope you guys enjoyed this video if you have any questions about what to do and where to get started leave them down below and I'll definitely try to help you out with that being said if you guys enjoyed please make sure you leave a like and subscribe to the channel for more videos about machine learning and stuff

Channel: Tech With Tim

Views: 344,825

Rating: 4.9493642 out of 5

Keywords: tech with tim, getting started with machine learning, how to get started with machine learning, how to learn machine learning, get started with ai, start machine learning, machine learning getting started

Id: nl7kDPYD20A

Length: 10min 32sec (632 seconds)

Published: Sun Sep 29 2019