RUclips's machine learning algorithm recommended this video about books for learning about hands on machine learning because I've been learning about machine learning tools through RUclips videos recently
I bought the book when I didn't know jack about machine learning. I found it wasn't as useful to me back then, as it is now. Its more of a reference book than anything else. Its better to learn the math/stats that goes behind the algorithms you use first before diving into this book. Otherwise you'll end up just treating ML algos as a black box where you type things in and magically get metrics on your models performance
Well what would you recommend ? As a first step ? I'm trying to work with deep learning without having any background in machine learning. I'm learning python too btw
For me, the real advantage of this book was that it smoothly showed me how to use the libraries of code; it saved so much time over digging into documentation (or doing a Google search) every step of the way. Often I can know the underlying theory and still pull my hair out trying to figure out the proper implementation. I feel like this book saved me a lot of time and aggravation.
Yes! I am so glad someone else has given a BIG thumbs up for this book. This book has lots of Python code along with some "easy digest" theory/math. I have a PhD in Comp. Sci. with nearly 20 yrs of industry experience and in a lead role. BUT, I still find myself in a situation that if I want to get a full rounded view of a topic - in this context, ML, I need to hit many other books: several in ML, several in Information Theory, several in stats, etc. Doing this iteratively is the key. In other words, it is VERY hard to write a CATCH-ALL book! So, tag along for a lengthy ride and you would get rewarded for your investment. I thank Aurelien for putting this book together.
I bought that book a year ago and yes, it's great. But I also bought 'Deep Learning with Python', by François Chollet, the creator of Keras (Manning Publications) and, in my opinion, this one is even better on Deep Learning. Both, in my opinion, are 'must have' books.
I also highly recommend the Deep Learning Cookbook by Douwe Osinga. He has great projects in this book and you can use them as a template for creating your own projects.
I watched this video, 24 hours later I had the book delivered. So far it looks good. The language is clear and the concepts are clean. I have yet to do any practical work this it yet. Thanks for the recommendation!
Wow, that was more reflected than I thought :-) Criticizing Aurelien for being too practical is like criticizing Ian Goodfellow for being too theoretical. Both books serve different purposes and (in my opinion) should be read as complements.
Dear Giles, thaks a lot for your channel! I bought this book about 6 months ago, due to a recommendation by Amazon. A little bit of my background: civil engineer, I have always used computers to solve problems, I have programmed in Basic, VB, Clipper, asp, C#, SQL and now, Python, as I found it was incredible to my ML needs. As an engineer, I could not agree more with what you say in this video. this book is incredible if you want almost instant, hands-on solutions to a problem that you decided to apply ML. Regards, Carlos.
Hi Carlos , I'm from mechanical background , but have landed a job in IT. Right now I'm interested in Data science and machine learning. Learnt theoretical stuff. Programming skills are c, Sql basics, python basics. Is this book still useful ?
I'm a software engineer and a mobile app engineer who want to explore the ML and AI sea, and after a few courses (wich I didn't finish), and learning Python (wich I felt in love with it) this book is the best book I ever seen for learning ML, I wich just to have time to finish my courses and read it from A to Z and try to do some dope projects. And for everyone who want to learn ML good luck and have fun :)
I'm halfway through this book and indeed it's really great. That said, you will benefit from it fully only if you have already fairly strong background in Python and you know most of the theory behind the algorithms already. But it's definitely very practical.
Currently going through this book. Love the content. I agree that the theory is not all there at times, and it makes assumptions that could leave you confused, but it's a great book at getting your feet wet with all the various topics in ML. It introduces concepts just enough to give you a good understanding, but for a deeper learning experience, I would supplement this book with some of Josh Starmer's videos.
Hello there, I am thinking of doing a RUclips video series on the book. If you're interested, let me know. Basically, the series will cover the contents of the book.
Thank you for this great review, I'm somewhat new to Python, and love the language and have been interested in Machine Learning, your rave review was SO effective, I didn't even finish the video, I went straight to the link you provided and bought it immediately. From the comments below it sounds like many agree with your thoughts, now I can't wait to get that smiley box in the mail :) Thank you kindly, I appreciate you and your wonderful channel.
the second edition has actually been recommended by the Tensorflow team multiple times personally however I like PyTorch more and even Tensorflow's own team has switched its API to look more like PyTorch.
It's rare that I leave the comments on YT but... This book is good if you want to plug-and-chug. What I mean is that, it will teach you what you need to APPLY ML. However, applying ML algorithms are easy. The hardest part is data pre-cleaning and trying to interpret the outcome of a model.
I was watchig a lot of sentdex's videos on machine learning, coupled with this book, I got a handle on the whole subject in under 2 weeks..power of youtube
I read the first chapter now and i like the book. It is written very clear, without pointless and irritating informations. Although my english skills are at german school level I had really big fun in reading.
I was using this book during my Machine Learning course this semester. It is really comprehensive. I had no idea what the actual process of developing a ML classifier or regressor involved beforehand and the book made everything really clear and made me even more curious about this subject.
so true. this book is epic to someone who wanna try to learn machine learning including real coding. obvious that the writer is concentrating on important subjects but he covers all around ML world well. I recommend this book to anyone who wants to try ML. Keep it up!
But the guy that wrote the book is French :-). The French are on average still pretty good at math, I think it has to do with their schooling & academic system, I believe it is quite brutal.
Thank you. I re-watch this video again. I used to find that book unorganized... and I just read the content page again. I missed chapter 2 data cleaning after your reminder. We can't do any sklearn without cleaning. I will read it now.
I picked this book to start with machine learning and I have to say that it was a big mistake. I didn't really understand anything as it is and, I can't believe that I am saying this, too practical. I was missing vital background information/theory in order to even understand the most basic concepts of machine learning. Sure you could just work through the book while looking up everything that you don't know but in my opinion, this is not a good approach. To me personally, first doing Andrew Ngs course was much better. I haven't really touched the book ever since but I could imagine that picking it up right now would make much more sense. Especially for machine learning, in my opinion theory matters. Especially when it comes to optimizing your models.
It's 2019 so this book is a couple of years old. But Amazon informs me: "There is a newer edition of this item: Hands-On Machine Learning with Scikit-Learn and TensorFlow 2e $97.01 This item will be released on August 4, 2019."
**Do not buy this book yet....** I am more into research aspect of AI, but still I bought this book a month ago... And I can confirm it's a one of the best book for doing practical ML stuffs and not only it gives you knowledge about the tools but it also gives you enough intuition of how each functions works, which is essential if you want to understand what those scikit-tenserflow parameters do.... But don't buy it for now, wait for the next edition to come, because a lot has changed in in the new version of tensorflow! So, wait for a little while Have a nice day :D
For anyone who is already an ML practitioner, I wouldn't recommend this book. If you are already comfortable with the higher level concepts then I would recommend the Elements of Statistical Learning or Learning From Data.
I've gone through the book the last couple of days and it's an absolute masterpiece! Thank you a lot for the review, but could you please consider making a top mathematically-oriented and rigorous machine learning books video? I know you've reviewed some in the best machine learning books video but I'd love to have your opinion concerning this. Thank you in advance and keep going! :D
Introduction to Statistical Learning is the best one out there although the code part(present as implementation at the end of each chapter) is written in R(which is not a problem as you can get the code part in Python from Github). However, the book assumes you have a reasonable understanding of Statistics before jumping in.
@@viacheslavspitsyn2995 I have never used Bishop's book though I have heard about it and read it's reviews. Here is my two cents based on reading reviews from different places: This book is like the Bible of ML but is very Math intensive and hence whether or not you should read the book totally depends on the level of Mathematics you are comfortable with. At few places I read even Math PHDs saying that they found this book hard to comprehend, this is no way mean to discourage you but I am just stating what I observed.
i would rather say ' once you know the theory', read it, your project will be implemented quicker... 1 page to explain svm and the vapnik dimension, that's great; one page to understand kernels, great as well; one page for a pca, great again you don't need theory, as long as everything is ok in your project and in your data........ well, on the other hand, if everything is ok, there is no need to pay for a datascientist people just reading this book should developp a recureent neural model to trade stocks, and... trade ( and let us know what the result is) cheers
It's a great book, but many of the library commands have changed, which breaks the examples. Best to follow using the corrected code on Aurelien's website.
I recommend you know linear algebra, statistics and probability, in addition to knowing how to program with the python language, and if you want you can learn R, if you already want to get into the neural networks part, I recommend knowing differential and integral calculus, the latter is quite complex and I recommend that you be patient because they are university subjects, they are not the most complicated in mathematics but they are quite difficult, excuse my English, I am just beginning to learn. Regards
Thank you for the overview! I will need to check it out! I just created a RUclips Channel and uploaded my very first video! It is based on Tensorflow and my journey to learning machine learning! Feel free to watch and comment! I wish everyone a splendid day!😊👍
Although the table of contents is covering wide topics on ML, the individual depth on each of these topics is still quite shallow for solid machine learning understanding. This book could serve as a python premiere for ML after acquiring sufficient theoretical background from books such as ML a probabilistic perspective or the pattern recognition book.
Want to let people know that there is a 2nd addition coming out in August which will include all the new stuff in Tensorflow 2.0 and Keras. I am waiting for that one.
Another video:
Should you really do 10000 steps a day? ruclips.net/video/gY8dom3TwMY/видео.html
RUclips's machine learning algorithm recommended this video about books for learning about hands on machine learning because I've been learning about machine learning tools through RUclips videos recently
now thats awesome
Machine learning all the way down
Because RUclips knows that Books > Videos
Sounds like a recursive algorithm.
The only thing is that the youtube algorithm doesn't comprehend that by educating more people in the ways of AI will assist its own potential breakout
I bought the book when I didn't know jack about machine learning. I found it wasn't as useful to me back then, as it is now. Its more of a reference book than anything else. Its better to learn the math/stats that goes behind the algorithms you use first before diving into this book. Otherwise you'll end up just treating ML algos as a black box where you type things in and magically get metrics on your models performance
Moby y can you recommend how to learn machine learning
A comment did a better job than the whole video, I feel the same I am in CNN's rn
i agree with you
Well what would you recommend ? As a first step ?
I'm trying to work with deep learning without having any background in machine learning.
I'm learning python too btw
Beautiful comment. Well done.
For me, the real advantage of this book was that it smoothly showed me how to use the libraries of code; it saved so much time over digging into documentation (or doing a Google search) every step of the way. Often I can know the underlying theory and still pull my hair out trying to figure out the proper implementation. I feel like this book saved me a lot of time and aggravation.
I was struggling through everything on my own, and my 2nd edition copy has just arrived. Much psyched! Thank you for recommending this!
I highly recommend .
Yes! I am so glad someone else has given a BIG thumbs up for this book. This book has lots of Python code along with some "easy digest" theory/math. I have a PhD in Comp. Sci. with nearly 20 yrs of industry experience and in a lead role. BUT, I still find myself in a situation that if I want to get a full rounded view of a topic - in this context, ML, I need to hit many other books: several in ML, several in Information Theory, several in stats, etc. Doing this iteratively is the key. In other words, it is VERY hard to write a CATCH-ALL book! So, tag along for a lengthy ride and you would get rewarded for your investment. I thank Aurelien for putting this book together.
I am Just Started Learning ML. Please Help me On this ML. Which Book is Best To Learn ML.
I love this book. I bought it last year. I recommend it as well.
This book use python for ml
I am a complete beginner in this field so, should I use it?
I bought that book a year ago and yes, it's great. But I also bought 'Deep Learning with Python', by François Chollet, the creator of Keras (Manning Publications) and, in my opinion, this one is even better on Deep Learning.
Both, in my opinion, are 'must have' books.
I've recently started this book and I'm loving it! It's really intuitive. Excited for the second edition :)
How's it.... I m mechanical engineer. And love maths.
Dude can u please state the prerequisites required to start learning from this book.
The beauty of well-spoken English language.
I am not a native english speaker, and i can understand him perfectly
Exactly ❤️
I like the book and I like the way Giles explains. You can see he's very passionate about Machine Learning and sharing his knowledge with us.
I also highly recommend the Deep Learning Cookbook by Douwe Osinga. He has great projects in this book and you can use them as a template for creating your own projects.
Such a well articulated review, I just ordered and subscribed.
i'm in love with this bookshelf
King
This is totally the best book I've ever encountered in the world of Machine Learning and Data Science
Aurelien Geron also has a youtube channel. He explains Hinton's capsule networks way better than any other explanation out there.
I watched this video, 24 hours later I had the book delivered. So far it looks good. The language is clear and the concepts are clean. I have yet to do any practical work this it yet. Thanks for the recommendation!
I love your editing work and your videos they're so good it feels unscripted
Thanks sooooo much for the advice! bought it right away!
Wow, that was more reflected than I thought :-) Criticizing Aurelien for being too practical is like criticizing Ian Goodfellow for being too theoretical. Both books serve different purposes and (in my opinion) should be read as complements.
We are waiting you to explain the relations between Artificial Intelligence, Machine Learning, and Data Science.
Dear Giles, thaks a lot for your channel! I bought this book about 6 months ago, due to a recommendation by Amazon. A little bit of my background: civil engineer, I have always used computers to solve problems, I have programmed in Basic, VB, Clipper, asp, C#, SQL and now, Python, as I found it was incredible to my ML needs. As an engineer, I could not agree more with what you say in this video. this book is incredible if you want almost instant, hands-on solutions to a problem that you decided to apply ML. Regards, Carlos.
Thank you!
Hi Carlos , I'm from mechanical background , but have landed a job in IT. Right now I'm interested in Data science and machine learning. Learnt theoretical stuff. Programming skills are c, Sql basics, python basics. Is this book still useful ?
Narsinha Kulkarni Hi, Narsinha. I understand that if you want hands-on advice & technique, this is a great book.
I'm a software engineer and a mobile app engineer who want to explore the ML and AI sea, and after a few courses (wich I didn't finish), and learning Python (wich I felt in love with it) this book is the best book I ever seen for learning ML, I wich just to have time to finish my courses and read it from A to Z and try to do some dope projects.
And for everyone who want to learn ML good luck and have fun :)
I'm halfway through this book and indeed it's really great. That said, you will benefit from it fully only if you have already fairly strong background in Python and you know most of the theory behind the algorithms already. But it's definitely very practical.
Currently going through this book. Love the content. I agree that the theory is not all there at times, and it makes assumptions that could leave you confused, but it's a great book at getting your feet wet with all the various topics in ML. It introduces concepts just enough to give you a good understanding, but for a deeper learning experience, I would supplement this book with some of Josh Starmer's videos.
Hello there,
I am thinking of doing a RUclips video series on the book. If you're interested, let me know.
Basically, the series will cover the contents of the book.
I'm interested
@@saikumar-gc2yh What is your background? Do you know the contents of the book?
Thank you for this great review, I'm somewhat new to Python, and love the language and have been interested in Machine Learning, your rave review was SO effective, I didn't even finish the video, I went straight to the link you provided and bought it immediately. From the comments below it sounds like many agree with your thoughts, now I can't wait to get that smiley box in the mail :)
Thank you kindly, I appreciate you and your wonderful channel.
second edition is coming in June / 2019!
August I think
Thanks mate saved me $50 bucks .. will wait for it in amazon.
@@vchandm23 it worth each cent. Buy the new one. :)
the second edition has actually been recommended by the Tensorflow team multiple times personally however I like PyTorch more and even Tensorflow's own team has switched its API to look more like PyTorch.
Got to know anout this book from your review. Ordered it immediately, enjoying it so much! Thanks for the review
It's rare that I leave the comments on YT but... This book is good if you want to plug-and-chug. What I mean is that, it will teach you what you need to APPLY ML. However, applying ML algorithms are easy. The hardest part is data pre-cleaning and trying to interpret the outcome of a model.
I was watchig a lot of sentdex's videos on machine learning, coupled with this book, I got a handle on the whole subject in under 2 weeks..power of youtube
Man_V_Snacks I don’t think you’ve got any handle at all. What you’ve got is cursory knowledge
I read the first chapter now and i like the book. It is written very clear, without pointless and irritating informations. Although my english skills are at german school level I had really big fun in reading.
I will surely buy it! Thanks sir, for your consistent hardwork......
You're very welcome 🙂
I was using this book during my Machine Learning course this semester. It is really comprehensive. I had no idea what the actual process of developing a ML classifier or regressor involved beforehand and the book made everything really clear and made me even more curious about this subject.
What's your major
Physics
@@chicagoboy279 Software Engineering
I really love it when a youtuber actually tells you what he thinks is the best, not just stops at the point where he presented all the options
so true. this book is epic to someone who wanna try to learn machine learning including real coding.
obvious that the writer is concentrating on important subjects but he covers all around ML world well.
I recommend this book to anyone who wants to try ML. Keep it up!
I have this book and it's fantastic.
I absolutely love that bookshelf.
I have taken the Tensorflow course of this books Author. Trust me he is the best when it comes to Tensorflow.
Can you please tell have you taken his Tensorflow course. Searched on net but couldn't find it.
It’s a great book to learn while practicing. I highly recommend it !
Damm British reasonablity strikes once again !
But the guy that wrote the book is French :-). The French are on average still pretty good at math, I think it has to do with their schooling & academic system, I believe it is quite brutal.
It is one of the best books that I have ever read
Ordered my 2nd edition yesterday 👍
Thanks for everything in the description too
Thank you. I re-watch this video again. I used to find that book unorganized... and I just read the content page again. I missed chapter 2 data cleaning after your reminder. We can't do any sklearn without cleaning. I will read it now.
Just ordered the new version. Thanks for your review!
Have new version came?
It is the best book. I’ve used it to pass ML class
Mo Azim which class you attended?
I picked this book to start with machine learning and I have to say that it was a big mistake. I didn't really understand anything as it is and, I can't believe that I am saying this, too practical. I was missing vital background information/theory in order to even understand the most basic concepts of machine learning. Sure you could just work through the book while looking up everything that you don't know but in my opinion, this is not a good approach. To me personally, first doing Andrew Ngs course was much better. I haven't really touched the book ever since but I could imagine that picking it up right now would make much more sense. Especially for machine learning, in my opinion theory matters. Especially when it comes to optimizing your models.
It's 2019 so this book is a couple of years old. But Amazon informs me: "There is a newer edition of this item:
Hands-On Machine Learning with Scikit-Learn and TensorFlow 2e $97.01 This item will be released on August 4, 2019."
For a second there i thought the lizard was 3D.
2nd edition is coming this summer. I have the 1st edition but I will buy the new version. The author will update the code with the new tf 2.0.
I love you and your channel, grateful for your content and guidance! I'm going to get this book 👌Thaaaaanks
I love you enthusiasm , Getting the book , keep up the good work !!
Your*
The newer edition comes out on June 30th, which includes *keras* on top of the scikit and tensorflow stuff
Really?
Excellent Cinematography style, keep the audience attached. Great work.
This was recommended to me because I clicked on O'reilly link under description section of one of the videos that I watched recently.!!
Excellent Review .. Subscribed and now going to Amazon to order
Second edition available since october 2019, keras added...
also updated for tensorflow 2. Thanks for pointing this out gonna buy the second edition!
**Do not buy this book yet....**
I am more into research aspect of AI, but still I bought this book a month ago... And I can confirm it's a one of the best book for doing practical ML stuffs and not only it gives you knowledge about the tools but it also gives you enough intuition of how each functions works, which is essential if you want to understand what those scikit-tenserflow parameters do....
But don't buy it for now, wait for the next edition to come, because a lot has changed in in the new version of tensorflow! So, wait for a little while
Have a nice day :D
There's lots of changes in every version of tensorflow ;)
Thanks for your review I already have it.
Can’t recall having any bad O’Reilly book in my library.. Nice video, thanks! ✌️
Is it still worth a buy when the newer edition is coming out, updated with tensor flow 2?
Thank you for putting up this review. It's been really helpful :)
For anyone who is already an ML practitioner, I wouldn't recommend this book. If you are already comfortable with the higher level concepts then I would recommend the Elements of Statistical Learning or Learning From Data.
Learning from Data by Abu Mustafa?
I am wondering. In your opinion is this still the best book?
Other commenters opinions are also welcome :)
Thank you very much!😊
I remember back in 2017 trying to use batch normalization on tf 0.x and this book had the one goddamn line to make it work.
I've gone through the book the last couple of days and it's an absolute masterpiece! Thank you a lot for the review, but could you please consider making a top mathematically-oriented and rigorous machine learning books video? I know you've reviewed some in the best machine learning books video but I'd love to have your opinion concerning this. Thank you in advance and keep going! :D
Introduction to Statistical Learning is the best one out there although the code part(present as implementation at the end of each chapter) is written in R(which is not a problem as you can get the code part in Python from Github). However, the book assumes you have a reasonable understanding of Statistics before jumping in.
@@sourabhkhandelwal689 What do think about Bishop (compared to ESL)?
@@viacheslavspitsyn2995 I have never used Bishop's book though I have heard about it and read it's reviews. Here is my two cents based on reading reviews from different places:
This book is like the Bible of ML but is very Math intensive and hence whether or not you should read the book totally depends on the level of Mathematics you are comfortable with. At few places I read even Math PHDs saying that they found this book hard to comprehend, this is no way mean to discourage you but I am just stating what I observed.
Hey Sir! Thanks for that review. I hat lot of doubts about that book and u make amazing work with this video!
What a interesting introduction 😹 I read this book too. It is really an useful book in machine learning 👍
I know it's great book Sir.....thanks v much for recommendation ..👍I have read a lot about it.
You're welcome 🙂
Thank you very much sir for your recommendation
It's one of my favorite books in the field! But the second edition is coming out soon, so wait for that. :)
Seems like a really good book. Best thing as you said is that it's practical. I wish I could buy it!
Out of the 63 books that I have about programming, Hands On Machine Learning with Scikit Learn and Tensorflow is one of them.
Grokking Deep Learning - best book
You didn't said anything about the prerequisites for this book?
Always love your book reviews
Thank you. I am a Java developer trying to learn ML
i would rather say ' once you know the theory', read it, your project will be implemented quicker...
1 page to explain svm and the vapnik dimension, that's great; one page to understand kernels, great as well; one page for a pca, great again
you don't need theory, as long as everything is ok in your project and in your data........
well, on the other hand, if everything is ok, there is no need to pay for a datascientist
people just reading this book should developp a recureent neural model to trade stocks, and... trade ( and let us know what the result is)
cheers
The 2nd edition is even better as it covers Tensorflow 2.0 in detail.
It's a great book, but many of the library commands have changed, which breaks the examples. Best to follow using the corrected code on Aurelien's website.
I wasn't sure from your video so I thought I'd ask: would you recommend this book?
lol!
I was about to buy this book but I recently read that there’s a new version coming up so.. i think I’ll wait for it 😅
When it will come... ?
On their website it states July, so soon :)
i read this one and yes it is one fine book
Ankit Gupta is it good for beginners
I would like to know is it good to beginners
I have done the python and math segment , can i now ready to jump to this book
Nice, I have just finished part 1 of the book before watching this video :D
I just ordered this book, I hope I don’t regret it 😅
Thank you very much for suggesting
I like how you put your videos togeather
Thank you. I bought the book from amazon just now
You are doing a great job.... Keep up the good work.
How much difficult is artificial intelligence? Should I learn ML first? What do I need to AI?
I recommend you know linear algebra, statistics and probability, in addition to knowing how to program with the python language, and if you want you can learn R, if you already want to get into the neural networks part, I recommend knowing differential and integral calculus, the latter is quite complex and I recommend that you be patient because they are university subjects, they are not the most complicated in mathematics but they are quite difficult, excuse my English, I am just beginning to learn. Regards
Honestly all O'reilly books are amazing my favorite is C# cookbook
Hi, great video and great recommendation. Would you recommend learning the theory before going to this practical book? Or
does not matter?
Thank you for the overview! I will need to check it out! I just created a RUclips Channel and uploaded my very first video! It is based on Tensorflow and my journey to learning machine learning! Feel free to watch and comment! I wish everyone a splendid day!😊👍
Although the table of contents is covering wide topics on ML, the individual depth on each of these topics is still quite shallow for solid machine learning understanding. This book could serve as a python premiere for ML after acquiring sufficient theoretical background from books such as ML a probabilistic perspective or the pattern recognition book.
2nd edition will be released this year.
I absolutely agree with you Sir. I have this book too. A Fantastic Book. Jesus bless you. Many thanks from Indonesia Archipelago. 🙏🙏🙏
Want to let people know that there is a 2nd addition coming out in August which will include all the new stuff in Tensorflow 2.0 and Keras. I am waiting for that one.