Bourke Brothers hit the library | Learning Intelligence 41

Check out that leverage. Where are we off to this morning, boys? Library. We're gonna get bean water and then we're gonna go to the library. Bean water is code word for coffee. You know, they put coffee beans in the water and it tastes great. But then we're off to the library to do some study. We've got the whole clique today. Josh, what are you studying? I'm going to study some more Core ML on Swift. Core ML, Swift. Billsy, what are you on? Uh, do some Python first, just introduction to Python, and then do some data science. That's what we're all about today. Back to— not to basics. I'm Python to data science. As well. That's what it's about, just repetition, right? Reps, reps, sets, reps, sets, reps. If anything I've learned from the gym, that's how to get good at something, just continual practice. Rip it out. You know how they get that shot where they're like this? Am I doing it right? You got to hold it out front. There we go, we've nailed it. How's it taste, bro? A good bean should. Caffeinated. Mr. Car. Phone away. Me? No, I'm talking to me. This is a quiet room, but it echoes. Echo! Hello, have you booked the room? Yeah, for Daniel. Daniel, so you were in— meant to be in room 3? Yeah, but there was someone there when we got here, so we figured we'd take the other one. Just see who's in that room now. No worries, thank you. Alright, hit a roadblock. Time to take a break. I'll show you. So this is, this is one of the biggest assignments I've ever done, including all the nanodegrees. Look how many tabs I've got open. All these questions, all these different forums. We scroll back up through. Look at this. What is it? It's more pandas. We go down, error codes, error codes, error codes. Where are we up to? There, I've got about 3 notes on what I need to do next. You can't even read here. Create continent column by mapping. Anyway, I've got some next steps on what I need to do. My brain's fuzzled. We're gonna head over to gym and solve it in the subconscious and then I'm gonna come back later this afternoon and finish off the assignment. But Josh has built something cool, check this out. What do you got rolling, Josh? This is basic at the moment. This is an identifier type app using an ML database that's already preset by Google. They have a, like, it's called Inception V3, I think. With the app, you take a, you click on the photo, the camera button, it brings up the phone's camera. Only works for iOS, but take a photo of what you want, say this water bottle here, and then you click use photo, and then it will read the photo through the database. And it thinks, like in the database, not that big, but it thinks that's a wine bottle. That's the highest. So you've built an image recogniser using AutoML— no, ML— CoreML? CoreML. CoreML. That's awesome. It's preset, so you don't need internet or anything. The, uh, the database is in the phone with the app. If we put a trained model in there, we can— it's not that well trained, but if we put something that's more well trained— no, but I mean, if we put our own model in there, CoreML can use it on the phone locally and recognize images. Yes. Ooh, I've got some ideas. Keen to work out? Yeah. Bills, do you work out? How was your prog? Uh, it was good. Uh, I'm up to nearly finished the first introduction course on Python and then go back to some data science after you finish the assignment and help me. Let's lift! Would you look who it is! Hey man, what's up? Keen to work out? Yeah, pretty keen. Yeah boy! I realized in that last clip I was pretty cooked in explaining what I was up to on that problem because you know when you get so deep into something and then you You step away from it for a second and you try to explain it to someone else and you've just been learning it for so long, it's like, "I don't know what I'm doing." That's what I use having a break for, is letting the subconscious process the problem and then coming back to it and understanding it from a new set of eyes. Look at those rock-solid arms. Awesome. What'd we start off? We did some ring dips, superseted with some pull-ups, and then we kept it simple actually, all about the rings and all about body weight. You can reproduce this workout wherever you are with push-ups and pull-ups. And then after the ring dips and pull-ups, we went on to ring push-ups. And then some, some just, uh, what's it called? TRX. TRX bodyweight pulls. And they, they were amazing. Now we're off to get some food, fill up these empty bellies. It's what, 2:45? Haven't eaten yet. Yeah, look, sorry about that. 2:55, 3 PM actually. We're doing a bit of fasting lately. That's, that's how we eat. Usually don't eat before 12. Yeah, 12 till about 8 PM. It's an easy way to stay shredded and lose weight if you want to give it a try. You don't need to eat as often as you think. Oh, chicken, rice, done. All showered up. You know, I haven't had a nap yet, but that's alright. I'm too excited to get back into this problem. I want to solve this assignment. I want to finish it. I'm currently at 60 out of 100. I need 80 out of 100 to pass, and we're not leaving this room until I pass. You keep your eyes on me. You watch. Oh yeah, check it out, a little bit of progress. There we go, assignment 3. There was a submission today at the library, 60 out of 100, then again 67 out of 100, then again 74 out of 100. Let me just quickly show you what my workflow is for when I'm having some trouble. So see here, question 13: convert the population estimate series to a string with thousands separator using comments, do not round— comma, sorry, do not round the results. So, a.k.a., get this number, return upon that number. So what I'll do, first of all, I'll come up in another browser and I just literally Googled pandas separate column value thousands with commas. And one of the first results came up with this. The person is working on the exact same function I am, which is, which is kind of crazy, but that's the internet for you. And there we go. Boom. Now I take that, plug it into my function here, and it worked. But what I didn't do is copy paste it completely. I copy pasted the whole thing, but then I rewrote out the code to make sure I could understand it. Now I'm going to figure out what exactly this little function here does, because I'm pretty sure that does the brunt of the work for me. Back to it. Woo! How good does a good yawn feel? One of those ones where your eyes water, your mind clears up, and you forget where you just were. Nah, I'm kidding. Check it out. We finally passed the assignment after about, what is it, a dozen attempts. Have a go at that. I'm gonna leave it there. Not fully at 100 yet, but I spent way too much time on this. I think there's a conflict between the autograder. Let me just show you. The grader output. So, I've read a bunch of the forums and even though some of my answers are correct, some of them don't match up with the autograder. I think I've only got one wrong. Where was it? Anyway, 94 out of 100. That is good enough for me for now. We're on to week 4. Let's check out what's in week 4. Oh, hey, man. What's going on, bro? Sorry, were you filming something? Yeah, I just passed my assignment. Oh, fuck yeah. Hey, Turkey. Movie? Yeah, I was gonna ask, what do you want to have for dinner? Because I'm hungry. Yeah, I'm pretty hungry too. All right, let me show them what Week 4 is, and then we'll get on to movie and dinner. That's good. You want to check out Week 4 too? Nah, I'm right. Yeah. Here we go. Past the deadline. Oops. Anyway, in this week of the course, you'll be introduced to a variety of statistical techniques such as distribution, sampling, and t-test. Oh yeah, majority of the week will be dedicated to your course project. Yes, week 3's course project took me, I would say, at least 12 hours. It said 2 hours, but whatever, give or take 10 hours. This project is suitable for your data science portfolio. Oh yeah, probably I'll upload it to GitHub. Yeah, that's a good idea. Once I finish, hold me to that, once I finish week 4, I'll put up a version of it to my GitHub. And then what do we got here? Statistical analysis in Python. Project— what's this one take? 4 hours. Whoo! We're in for a big one there. So we've got introduction, distributions, more distributions, hypothesis testing in Python, end of theory. Beautiful. And then that'll be part 1 of the applied data science with Python. Oh, Daniel, I'm sitting in that chair. Oh no, no, no. Oh no! Told you I was. Hello guys, week 4. Anyway, welcome to week 4. Let's go get dinner. Shake it, shake it, shake it. Why would we watch a movie when we can just create our own entertainment? We actually did choose a movie. What is it? Some gangster one with Johnny Depp? Yeah, I don't know the name. It wasn't Johnny Depp, it was Tom Hardy. No, it was Johnny Depp. No, it wasn't. Anyway, check out my favorite dessert at the moment. Coconut yogurt, almond butter, and down the bottom there's mixed berries. No sugar, and it's absolutely divine. Would you want to wrap this up? Can confirm, absolutely divine. Yeah, can confirm, can confirm, can confirm. Look at that, 3, 3 yeses. All right, we're gonna sign off. 3, 2, 1. You both missed it. Keep learning. Yeah, Zed got it. We'll see you next time, team.