deeplearning.ai Progress! | Learning Intelligence 14

What's going on, learners? Welcome to Learning Intelligence Episode 14, and today has been a really fun day. I've been trying out this new learning platform that Google have released called Qwiklabs. And why is that? Well, because in a few days Google are hosting an event called Google On Air, I believe, and they— there's an event they host every so often where they have, uh, speaks and speakers and talks about stuff that's going on in the Google world, and this one is on The topic is big data to AI, and so one of the prerequisites for like attending the event or watching the event is that you have some basic idea of the Google Cloud Platform. So that's what I've been doing here. I've been learning the— where is it here— Google Cloud Platform Essentials. I went through all the modules today, 7 out of 7 labs complete, and it's— I must say it's one of the best hands-on lab experiences I've ever had. And I'll put a link to this in the description, by the way, google.quicklabs.com, so you can check it out. There's a bunch of free modules on this. You can try it out, use the Google Cloud Platform. If you ever used AWS before, it's similar to that, but I found the Google Cloud Platform to be much more user-friendly. So, that may be something you want to look into in the future to use for your applications or use for hosting a website or something like that. And the reason why I'm doing it is because I'm going to need to, to run my future applications on something, and cloud computing is where it's at at the moment, especially with Google and Amazon and, and all their all of their APIs that they have, their machine learning APIs. I'm looking to eventually take advantage of those and put them into some application so that people can use it and bring value to the rest of the world. But I wanna show you one more thing. I just had a shower and so I'm about to go to sleep. You probably saw how orange my screen was, it's 'cause of Flux in full effect. But this is my journal and every morning I write in the date and about a paragraph or so of just random thoughts. I start off with one truth. So today's truth was, "Humility is the essence of life." And I believe that was, that's a quote from a former Disney CEO. I can't remember his exact name, but that's where I read it this morning, it's a quote. So humility is the essence of life. And I don't know, I write those down so it reminds me that I can take from the best and learn and try and apply these things in my own life. But what else is on here? I put in 6 goals for today. So write, I still got to do that actually, that's gonna be my next step. Workout, completed that. Google training finished, 1 Pomodoro Python, 2 Pomodoros on Learning Intelligence 13, so editing it, and a vlog clip, which is what we're doing now. So, I'm going to do a few more points in the journal before I go to sleep. Otherwise, tomorrow, I think I'm going to start the deep learning course on Coursera. So, we'll check in there. Check it out, guys, we're on the deep learning course on Coursera today. Just started it. About 90 minutes through the first module, and so far it's been amazing. I just had a break, I went for a walk to break up my study schedule, and so now I'm going to get back into another 19-minute block and then I'm going to call it for the day, but I'll check back in with my favourite parts so far. Alright, so if you've ever done Andrew Ng's machine learning class on Coursera, the deep learning specialisation takes all the good things, all the good teaching mannerisms from that and goes above and beyond. I don't know whether it's because I've got a bit of foundational knowledge in deep learning now and this is sort of reconfirming all that knowledge that I've learnt through things like the deep learning Nanodegree and various other online resources. But so far, I'm about 3-4 hours into the course and it is incredible. That's, that's going to be enough for today. But I've written down a few things that, that I picked up that were my favorite from it. So let's go through. First of all is from the welcome video with Andrew, and his passion for the AI as a topic is incredible. And he believes that AI, or his philosophy is that AI is the new electricity. So if you imagine how much electricity changed the world when when devices, when objects could suddenly have power to them. That's the same thing that's going to happen with artificial intelligence, or well, that's what he thinks at least, and I, I, I tend to believe that as well. If you imagine how many things are being connected to the internet these days, from, from these, from tablets to water feeders to plant feeders to everything, everything's being connected to the internet. And when you have a connection to the internet, you can suddenly start to stream things like intelligence, and that's That's where AI is going. Suddenly things that were not intelligent can have intelligence injected into them and become intelligent, and thus our interaction with them becomes deeper and deeper. Number 2 was the interview Andrew did with Hinton, or Geoffrey Hinton. And Hinton is quoted by Andrew as being the godfather of deep learning. He's such an interesting individual. I didn't know that when he graduated university, he did an undergraduate in psychology or something like that. He was always interested in how people think and how the brain works. And he left university, went and got a job as a carpenter for a bit, and then came back and did a PhD in artificial intelligence. So if you're on a break from studies or whatnot, if you're doing something, just remember one of the godfathers of deep learning went and became a carpenter before he even was interested in AI. And his philosophy on AI was much, much aligned with Andrew's. And he had some words of advice on anyone who's looking to get into it, is that read enough of the literature so you understand the concepts, but not too much because then you'll start to think like how the industry already thinks. So if you want to change things, read enough so you understand the fundamental principles and then start your own work, start implementing your own point of view on it. So number 3 is a little simple one and number 4 is simple as well. Is I didn't, all this time I've been using logs in my life, like logarithms, and I didn't know what an actual logarithm was. It's funny how you miss these certain concepts in math and they, you sort of, you can use math as a language, but you don't necessarily understand what the language is, what the language means. Like when you use a word in English, but you know the context and that it's relative in that certain context, but you don't exactly know the definition of the word. And so I googled what a logarithm actually is because it was used in some of the calculations we were doing. And a logarithm is essentially what multiplication is to division. So if you imagine, multiplication is the inverse of division, logarithm is the same thing but for exponentials. So if you, we take the number 1,000, so 10 times 10 times 10 equals 1,000, that's 10 to the power of 3. And so 1,000 to the log base 10 is 3. So it's essentially just the inverse of the exponential. Fun fact about logs. I'm up to my first assignment in the deep learning Coursera course. It's Python Basics with NumPy, and it says it'll take about an hour or so to to go over. And although it says it's optional, I'm going to go through it anyway because you can never have enough Python or NumPy. It's about halfway through the actual programming assignment of week 2 of the deep learning course on Coursera, and it's really fun, I must say. Like, this is, this is going on above and beyond. This course is amazing. And it's later at night. I don't usually code like till this late at night because I'm getting tired. But essentially what we're doing is we're building a neural network. Where is it? Ah, there we go. That's the track I'm up to. So I figured out if I put my— put myself to it, I can do about a week's worth of lectures and programming assignments per day. My goal is to have the entire deep learning specialization done before Christmas. So let's see if we can do that. But the project I'm working on is we're building a logistic regression with a neural network mindset. So essentially designing a neural network to decipher whether a photo is cat or non-cat. So for those who've seen it or, or heard of it, Silicon Valley is a TV series, and the guy on there built an app called hot dog versus not hot dog, and we're doing the same for cats. So cats are the animal of the internet. So what better way to start off learning about neural networks? Oh, by the way, there'll be a link to, to Call Sarah Search for the deep learning specialization. I did partner with them as an affiliate, so I'll put my affiliate link in the description if you want to sign up to Coursera to do the course that I'm doing or do any of their other courses. They do have 1,600+ courses on there, so definitely worth it. And if you go through my link, full disclosure, I will get some sort of kickback from it, but this is what I'm trying to do. I'm trying to bring value to the guys through making these videos, and then if I can make something to sustain my way of living through these courses, That's how I plan on doing it. Made some progress today, guys. I just passed the shallow neural networks quiz. So this is week 3 of the deep learning course. That was my 3rd attempt, by the way. So don't let a perfect score get in the way of someone failing the first 2 times. Well, at least that's what I did. I got 7 out of 10 on the first attempt and then 9 out of 10 on the second attempt. And then the 3rd attempt, I got 10 out of 10 because I knew what my previous answers were and I sort of just went off that. But I was pretty happy with my first attempt of being 7 out of 10 because some of the concepts that went through in this, this week were fairly math heavy. But the explanation that Andrew Ng gives is incredible. He goes through the fundamentals of what you need to know for things like forward propagation and back propagation with deep learning, and then gives you sort of the minimum effective dose, or the, yeah, that's it, the minimum effective dose of what you need to actually implement these algorithms. So don't go fretting if you're not sort of a wizard at mathematics. I've only had a high school mathematics education and then I've upskilled myself on Khan Academy and just watching various videos and reaching out to people for help. So don't worry too much about the math. Andrew Ng is a great teacher. And second of all, this— every time I go through this course, it keeps getting better and better. So I'm up to the second programming assignment so far. And as I said before in the previous clip, I'm looking to finish the entire specialization before Christmas. So, let's do it. Now, I'm not sure whether or not I mentioned it in a previous clip, but a couple of days ago, I attended an online event by Google Cloud, and it was called Google Cloud OnAir: From Big Data to AI. And there was some amazing lectures, some amazing talks from people who are really high up in terms of the Google Cloud platform and really great at what they do. And so they were sharing their knowledge of how the Google Cloud platform can be used and how it's been used already. But I had a few favorite takeaways. My favorite, actually, yeah, let's go through what it, what was it? A couple of them. One of them was the access that we have to this platform. So first of all, we have access to the production-grade, like artificial intelligence APIs that Google uses in their products on their free tier. So I'll put a link in the description or somewhere here. It's cloud.google.com/free. You can check out all the free stuff they have there. I won't go through it all in this video, I'm probably gonna make one, a video in the future going through some of the ones that are my favorite. For example, you can use their speech-to-text API, AI API. So say for example, you built an iPhone app and you wanted it to summarize your meeting notes. You can have that iPhone app link up to the Google Cloud API, set your iPhone on the desk, and it will use Google's production-level speech-to-text algorithm to convert the words that you're speaking into text. And that's something, not specifically that example, but that's, that's something that I want to try in the future is, is go onto the Google Cloud Platform and use their APIs to build some sort of application that, that can be used in everyday use. So if you're interested in that, I'd definitely check out cloud.google.com/free. That's their free tier. Otherwise, check out the Google Cloud channel on YouTube. They have a lot of great stuff there. And as I said, as I begin to use it more and more in the future. I'm sure Amazon will have a similar offering, but to me, Google is, from what I've used AWS in the past and what I've used Google briefly, Google's a lot more user-friendly. And I think that's something I value as really important. So I'll make sure I bring it out to you guys or show you guys if I'm using Google Cloud Platform for anything in the future. But that's enough for this video. It's time for some shoutouts. First of all, thank you to Pooja, Brendan, Zakia, Albert, Leonardo, and Sayem. I apologize if I pronounced your name wrong, but thank you guys so much for leaving your comments, reaching out to me via email. I hope the answers that I gave or the replies that I gave were helpful in some way, shape, or form, and if not, you can always follow it up, let me know if I can do better. And that goes for everyone else, actually. If you want to reach out to me at any time, my email is daniel@mrdbourke.com, and reach out to me with any questions about AI, fitness, food, life, whatever you want to talk about. I reply to all emails, and if I don't, send me a follow-up. And thank you Thank you so much to my latest Patreon, Jack. I really appreciate it. And guys, if you want to become a Patreon and further support this channel, the link's in the description. But otherwise, if you want to see anything in future videos, you have any advice for me, leave a comment below. I'll do my best to get to them all. Next week's video should be something about— should be me keep going in the— well, I'm going to keep going in the deep learning course on Coursera, and we should be definitely well and truly almost at the halfway or just over the halfway point by then. As always, Thanks for watching and keep learning.