Computer Vision Basics + More deeplearning.ai Progress! | Learning Intelligence 16

Key Takeaways

whoa check it out guys just past the second course on Coursera is deep learning specialization improving deep neural networks hyper parameter tuning regularization and optimization and look how beautiful that certificate is and that's something I can put on my LinkedIn or other profiles and whatnot you're gonna have about four or five of these five there you go by the end of this course so welcome to learning intelligence 15 this week or this episode is going to be all about the deep learning specialization on Coursera I just passed calls too and tomorrow I'm going to be getting into two course three which is on I think improving machine learning models but we'll get into that tomorrow I'll show you that otherwise the last project of of part two was on tensorflow and now I got a special guest who this episode so I'm gonna go try and find him and then he's gonna explain or exactly tensorflow is whoa I cannot believe that worked Shh don't so Daniel over here but I'm from the Year 2045 you could consider me a future dan and I just found a time portal I know I'm talking like a Batman but let's say the future is great and I have a explain tensorflow for you so if you imagine tens of love is it is a library it's a framework alright so if you wanted to build something out of Lego you don't build the Lego blocks yourself you buy lego set and you put it together and you make something amazing so if you want to build a really advanced deep learning model like they do at Google you don't build it from the ground up when you have tensorflow available which is your Lego said and you can use the same library as Google tensorflow to build deep learning models it's like creating a lego feature out of your brand-new Lego set but you're doing it with your deep learning model o-p-s from the future all the rick and morty episodes are practically true those guys are like fortune tellers back to Dan alright so I couldn't find my friend and maybe tensorflow I have to be explained another episode but that's alright tomorrow I'm going to get into part three of the deep learning specialization I still wait I'm gonna finish up today about halfway through week two of three on the Coursera deep learning specialization check it out here what are we up to ml strategy - so this part of the course or course three of the deep learning specialization I'm going to put you down here it's all about improving your machine learning model so once you've sort of built it and once you've hit the target of or got some good results at least or maybe you've got some bad results this week is teaching us all about how to improve those results or how to take the results you've got from hitting the wrong thing and adjust them so that they can be moved more towards the thing you or towards the answer you're really after actually one of my favorite takeaways from this so far has been that I know there's there's a lot of content on sort of look at this actually I'll show you my notes on a lot of technical stuff on how to improve your models what not doing test evaluating ideas and parallel whole buttons your quiz notes there there's look there's a quiz question that I got wrong and then we go up here I can't even pronounce this almost or ortho GaN elevation orthogonalization comment below if you can pronounce that better than I can so there's a lot of a lot of in-depth stuff and a lot of stuff that you can definitely definitely use to improve your models but one of my favorite parts of it was the interview with Andre capacity if you haven't heard of one breaker Pepe he's a computer science PhD who taught a course at Stanford on computer vision I'll put a link below to his his course actually his his insight he's currently head of AI at Tesla and one of his insights that he said at the end towards the end of the interview was that when they were trying to he thinks the through here today arguing on two different directions one will be more so building building models for different tasks and the models keep getting better and better and better and there'll be another sort of another branch off of AI which will be the team's working towards artificial general intelligence and what is what his insight was was he said that when they were first getting into computer vision and he's definitely an expert on computer vision he taught a course on earth at Stanford he said they were trying to do it in like compartmentalized pieces so for example they do facial structures and then do maybe animals and then do cars or something so different different subsets of computer vision tasks and I may be getting that wrong but just imagine different subsets rather than doing it as a whole piece and he thinks that's that's maybe the key to two other features of general intelligence is rather than nailing natural language nailing speech nailing computer vision and then trying to merge them all together it might be better to work on it as a system as a whole that was really cool for me actually there's a blog post let me find it and I'll show it to you quickly because I'm gonna read it and I'll link it in the description so you can read it too and comment below what you think so here it is a short story on AI a cognitive discontinuity gets I can't speak tonight so uh just just pretend I pronounce that word correctly but this is entendre Pappy's blog I'll link this below I'm gonna read it and I'll let you know what I think that's if it means our study tonight I'm actually I'm not gonna read that because I don't like looking at screens while I'm in bed this is the book that I'm reading at the moment it's incredible absolutely highly recommend it to short read it's only about 150 or so pages but Viktor Frankl was actually a prisoner and or switch and some of the insights in this book are incredible so if you're looking for a new book to read I check this one out so me I think 9 doubles or something on Amazon or something like that or Book Depository but I'll leave a link in the description so you can check it out and if you have read it come on below what do you think I want to show you possibly one of my favorite parts of this course so far so this is a quiz or actually it's a case study so at the end of week three or the no sorry course three and it's an autonomous driving you can imagine it as a flight and refers to it as a flight simulator for machine learning and as you can see I failed it the first attempt so got 11 out of 15 but the beauty is I can try it three times every eight hours so this is anima striving case study so to help you practice strategies for machine learning and this week we'll present another scenario and ask you how you would act we think this simulator of working in a machine learning project will give you a task of what lending of what leading machine learning project could be like I'm reading this through the camera screen Holloway and I should really just read it normally so essentially you are employed by a start-up building self-driving cars amazing you're in charge of detecting road signs stop signs for the turn signs construction signs ahead etc and traffic signals red and green lights and images the goal is to recognize which of these objects appear in each image as an example the above image or the below image it should be contains a pedestrian crossing a sign with red traffic lights so there we go example image example output so what the questions are is imagine you're in this scenario say for example I'm gonna show you one rather than me explain it let's go here first question would be you were just getting started on this project what is the first thing you would do assume each of the steps below would have take an equal amount of time a few days and I got this one right so the first thing you should do is spend a few days training a basic model and see what mistakes it makes so correct as discussed in lecture applied ml is highly is a highly iterative process if you train a basic model and carry out error analysis see what mistakes it makes it will help point you in the promising directions that's one example of the questions you go through and I'm really enjoying it because it's it sort of it really feels like as if I'm working on the self-driving car project it's like someone's coming to me and asking hey we've got this scenario what should we do here and of course I haven't got it all down pat yet because they only got 14 I mean 11 out of 15 I'm gonna do the quiz again in a second but I think it's really good to get this type of experience and as Andrew says he's I haven't he hasn't come across any kind of scenarios like this before in other courses and neither have I actually and so it's really good to get hands-on experience with what a case study would be like in in working in some sort of startup or some sort of company on these deep learning projects which is which is what I what I intend to do in the near future so that's all good an example of what I got wrong I got two wrong in a row for example your goal is to take road signs stop signs pedestrian crossing signs bunch of signs and traffic signals the goal is to recognize which of these objects appear in each image you plan to use a deep neural network with relu units in the hidden layers for the output layer a soft soft max activation would be a good choice for the output layer because it's a multitask learning problem true or false so I selected true but it should be false so maybe a softmax shouldn't be used for the the final layer I'll find out where I went wrong and catch up on the next attempt all right we made it so second attempt I got 100% because I went over my errors I took some screenshots at the questions that got wrong and and put them in here so that when I was going back over the question I could choose a different answer follow-up to the question I got wrong with all question two so I was actually false so softmax would be a good choice if one and only on the possibilities stop sign speed bump relation across in green light and red light was present in each image but we need more than one possibility softmax to be good if if this was red light traffic light was a zero and there was only one output but softmax doesn't help when we need more than in one potential output on to the next course so before getting in a course for I was watching this interview with Andrew and Ruslan salakhutdinov probably pronouncing that wrong but Ruslan is the head of AI and/or director of AI research at Apple which is really cool and it's amazing to get all these insights from some of the what's called its heroes of deep learning in this course and it's good to get their perspective and how they got into the field because know they all came from such different backgrounds like for example Ruslan did his master's in computer science and then he went and worked in the finance sector and then he randomly bumped into Geoffrey Hinton if you remember from in the last video the one before who's the Godfather of deep learning and then Hinton got him into deep learning and what he was working on an artificial intelligence and whatnot maybe not deep learning at the time but got him into AI and that's when he did his PhD fast forward a few years he's now head of a director of research a I research at Apple so that's amazing and I took a few notes on on what his favorite favorite parts of deep learning are and his advice on on how to get into deep learning one of his main takeaways was that when he teaches a class where is it yeah when Ruslan teaches a class on deep learning he practices he gets his students to practice creating a backpropagation algorithm algorithm for a convolutional network yourself so I'm excited to do that I'm doing that in the next course and he says deep reinforcement learning is very exciting trade training agents in virtual environments that's something I'm really excited about too I wrote an article on medium actually about deep reinforcement learning it's it's a high-level article so it's not it doesn't go in-depth on to what deep reinforcement learning actually is but it was on how deep mind trained alphago zero sorry Alfred going on Alfred zero and they've released a new environment for starcraft 2 to help train deep reinforcement learning agents but I'm on to the fourth course of the deep learning specialization which is on convolutional neural networks and we're going to start off with some computer vision so I'm gonna jump into this and then once I've made some progress I'll let you know alright guys so we just finished the first project of week one of course for on because Sarah deep learning on AI deep learning specialization and it was essentially coding a convolutional neural network by hand forward propagation and back propagation so we see all the way down at scroll look how small that scroll bar is so this was this is probably the hardest project yet in the whole whole specialization just because it was probably the most programming intensive one so far as well and the fact that convolutional neural nets if you'v they're kind of hard to understand if you haven't been over them before so what i've done is i've drawn a very basic overview or how I see convolutional neural nets so far and mind you I'm still a beginner so don't take this as gospel but this is how I I think about it and maybe if you check it out you could leave a comment and you know more about it than I do you could tell me how to improve on it here's what we have say a convolutional neural net by the way is used to recognize images so that's that's the main thing or video or images just imagine it for convolutional see for computer vision so that's what it's useful so imagine if we had a million pictures of trees i'm convolutional neural network starts off by making a little filter a little box and it's gonna go across this image will fill in the gap there and then it's going to keep going going going moving across and all the way through the image Philips done it layer by layer and then what it's going to produce over here is a matrix of zeros and significant values so if you notice the ones treat them as significant values it may not necessarily be ones but it could be something else so this means that the filter has picked up that there's something here in the image alright so the rest is all zeros and this is very roughly drawn drawn so forgive me for that and then it's going to convert it even further into eventually being a vector so it's going to be zero zero zeros and then some ones throughout there because after all what a computers understand best they understand numbers so a computer can't necessarily understand a picture of a tree so it turns it into numbers and then what do we do with those numbers well say we did that this process a million times right and from what from that what you would get is a whole bunch of these that you can use to recognize other pictures of trees what the computer could do is find patterns in those vectors so find patterns in the zeros and ones and then use those patterns to recognize future images or or scan over future images or images that you feed it and it will give back whether there's a tree in there or not because that new image will have similar significant features aka same amount of ones and zeros as other images you've fed it aka the millions of of images you've fed it as as the initial data and remember deep learning networks love a lot of data if there's ever one thing you want to do to improve your machine learning network or deep learning network or especially deep learning at work it's get more data get some good data make sure it's good mobile data because if bad data will make it worse of course and understand that computers don't understand what what a picture of a tree exactly is so or anything really for that matter the facial recognition on your iPhone 10 or iPhone X doesn't recognize your face at recognize it turns your face into numbers and it recognizes the numbers so that's the two main takeaways for deep learning as a whole in common neural nets and that's how I understand them if you have a better understanding of them please leave a comment below and help some other people out as I learn more about it I'll bring more to you after after this I've got to hold my next step in the deep learning specialization is another another project programming a silent program can't speak programming assignment but instead of coding it convolutional neural networks by a hand or step-by-step I'm gonna use tensor flow and as I've said before Pro deep learning libraries help take out a lot of the code so a lot of these functions are built into two libraries like tensor flow and carrots and that was a really long clip but we're gonna wrap up this video let's see I think it's the end of learning intelligence 16 we'll finish up with some shout outs and then I play into the next week by the way did you finish reading that story are linked I finished that story by Andre kept at me it was a really great read took me a little while probably about half an hour at least I won't spoil it for you it's definitely something as I learn more and more about artificial intelligence and and smart agents and whatnot in smart systems it's definitely something out I can see happening in the future all right shoutouts for the week these people reached out to me on various social media platforms Instagram Twitter all the links will be in the bio emails commented on YouTube and by the way you can email me at any time my emails daniel at mr deburr comm you have any questions if you have any emails everyone has emails if you have any questions about AI or or anything in general you just want to talk you want to meet happy to hear from you and I'll do my best and to it thank you Sabina Jake Gregory Kenny and Madison I really appreciate it guys thank you so much for reaching out it means a lot to me and so what's the place for the next week well as this video comes out I'm probably gonna be on an island so I won't necessarily be doing much my study at all I'm gonna be on holidays with my family and friends so I'm gonna take the week off and then starting early 2018 I'm gonna continue on with the deep learning specialization on Coursera what you've been seeing in the past a couple of videos and then of course moving through my AI master's degree so we checking off the steps there I think that AI nano degree term 2 which is based on deep learning starts on January 11 2018 so I'm really excited to that that's gonna be my main main thing or main main squeeze main squeeze is that even the right word to say probably not but oh well we're gonna roll with it main squeeze for the first few months of 2018 so happy New Year and thanks for watching leave a like subscribe if you want to and as always keep learning