I cannot express my happiness enough. You are inspiring many and I am going to strive to do as much as I can to make my students understand. True teachers are underappreciated. Much love.
This is the most crisp, lucid, and downright amazing explanation of the concept with demonstration. I could make out the efforts you have taken in designing this video, answering every detail of the concept and covering all possible scenarios every step of the way. You're hands down brilliant, thank you for this creation!!
Luis, you are a freakin’ genius 💜 There’s no way someone would not be able to understand any math concept if taught in such a lucid way. Sir, it’s a humble request from my side (and I’m sure a lot of others would agree with me) to create an ordered playlist on math for machine learning, it’d be of great help!
Thank you Debarchan! I have some playlists on the front page of the channel (ua-cam.com/users/LuisSerrano), on supervised/unsupervised learning, probability, etc. Also, I organized all the videos into topics here: serrano.academy. Let me know if they help, or if you have any different suggestions. Cheers!
So comprehensive yet always to the point. Amazing quality. My professor mentioned in passing that the Mars rovers perform SVD and only select the first few largest singular values to send back and you explained perfectly why this works!
Having to compose a mathematical essay about the usage of SVD in image compression, without having been taught about the concept of linear algebra, I had to look for appropriate sources that would help me understand the theory. After hours of online research I managed to find your video and I have to admit that I'm impressed. Your analysis is both thorough and comprehensive in the simplest way possible. Thank you so much for your help! Keep up the good work!
My professor taught this in class and I was like what is the use of this. After watching this, I realised how fascinating this stuff is. Thanks a lot for making this. Shared with my friends.
i have been finding study material to learn SVD and in this journey I went through several PDFs, slides and few books but this video is the best and crisp explanation of this algorithm. Thanks!
Two Thumbs Up!!!! I have been all over UA-cam the past two days watching presentations on SVD (many of them very good), and I am only 3:45 into this video and I **NOW** understand why there are two rotations involved with this simple illustration. This was gold and very, very helpful. Thank you!👍👍
Okay, mind blown a second time here at minute 13:00. Seeing the visual for why when the second scaling factor is zero it compresses the line and thus reduces the dimension via the degenerate transformation. Again, I've seen the math elsewhere, but not until the visual here did I actually understand what was going on. Thank you again.
I'm so happy to finally find you....the best...it was like you were answering my questions while I'm asking you in a live...hoping for more similar videos ❤️
Sheer brilliance. Your clarity of thought is mind bending. Luis serrano, statquest, ritvickmath, krish naik and codebasics - this is pretty much my holy grail for datascience
Easily the best explanation on youtube by far - I've been going all over trying to understand what SVD actually means. This video provided that and so much more, will catch your other videos and hopefully they're as good!
This explanation is genius. After watching a few videos on the subject, I got to this one and wow, everything I just saw made sense. A few people have this talent, and it’s awesome that you’re helping others with it. Keep up with your content!
This IS the best explanation of ANYTHING (no matter what field, what topic) I've ever got in my life about something. Incredibly visual didactic style. Grandioso!
You did a phenomenal job on this concept and video! This is the best explanation I have seen as to why we can drop smaller singular values and end up with a reasonable approximation!
You're an excellent teacher, Luis. Can't thank you enough for the amazing explanations for these hard-to-grasp concepts. Hope you and your family are doing well during this pandemic.
So informative and easy to follow. I love this. Thank you so much for taking the time to create this video. It's so important to know how the concepts we learn in class can be applied in real life. This has changed everything for me. Thank you again.
BRAVO! If anyone of you viewers wonder what "disruption" looks like when it comes to technology and business models, this video is a perfect example. If we just narrow it down to getting the knowledge of, say SVD, then why on earth would anyone pay a tuition fee, commute for hours or relocate, to sit in front of a mediocre "teacher" who do not have the interest or pedagogical skills to teach (we have all had these "teachers" and it sucks! Is that not why you are here?!?!), when you can watch a video of this quality with prefect explanations and examples at the comfort of your home as many times as you want for free(*)??? Eventually no one will!!! Education is truly democratized one video at a time...
I really love the way you explain these Maths topics which I never understand in Linear Algebra classes as theory but here I got in this one video with clear visualization and moreover we can always follow in python code which is really cool to see. Thanks for effort and time for strangers like me. Just Keep going.
Wow .it's so crazy and the mathematics representation was expressed in the pictorial representation of the beginning of this video is so awesome...keep rocking
Luis, gracias por compartir tus videos. Realmente son de lo mejor de YT. Siempre recomiendo que te sigan(te descubrí en Udacity)...me encanta la forma en que abordas SVD, es la misma forma que uso yo mismo: abordarla desde un punto de vista geométrico. Evidentemente, la calidad del video me deja sin palabras...la magia de SVD es la reducción de dimensionalidad y la muestras magistralmente, la clave de esta reducción la dices en 11:33 determinante = 0...¡Bravo!
Thank you! I came across SVD while doing natural language processing. This is the best explanation I have seen by far. Is there any chance you can do a chapter on SVD in the context of machine learning? Thank you again for the excellent work
What is your Patreon? This channel...if you keep going like this is much better than 3B1B. Perfectly done! That, Luis is a freaking awesome channel. Your’s is much better!! Do Variance-Covariance. Thanks.
Wow! A very much simple explaination of SVD with great intution and visualization! The best explaination of SVD I have found in internet so far. Thank you sir, also, could you please make a video about the computation of SVD in a detailed manner.
Great video. Suggestion: You discussed square matrix in details in the first 20 minutes, which looks more like eigendecomposition. SVD needs to be discussed truly for rectangular matrix for another 30 minutes. Your graphic presentation is the best I saw in youtube on this topic so far.
Yeah, you could go the projection route: get mean, center dataset, get x and y variance, calculate the covariance matrix. Get eigenvector/eigenvalue set. Choose the bigger eigenvectors that correspond with larger eigenvaluees. Project on lower dimension.
I am thinking from which point should I begin writing the notes because everything in the entire video seems so fascinating and resourceful that I cannot skip anything.
Thank you Kavya! I have one on matrix factorization here: ua-cam.com/video/ZspR5PZemcs/v-deo.html Haven't done latent semantic analysis yet, but there's one on latent dirichlet allocation: ua-cam.com/video/T05t-SqKArY/v-deo.html
After searching for this topic, I had a list of videos to choose from, then I remembered the great explanations I had gotten from this channel, so I selected this video... wasn't disappointed.
Pretty cool! I forgot a lot of my math studies in the last year, but your explenations were so precisely, but also simple that I understood nearly everything. Thank you (and sorry for my bad english :-) )
![SVD: Image Compression [Python]](/img/n.gif)
Best explanation on SVDs. Keep it up Luis !!
Thanks Ahmad, coming from a top youtuber like you, this is great praise!
I must admit one of the simplest and efficient explanation of SVD. Loved it.
I cannot express my happiness enough. You are inspiring many and I am going to strive to do as much as I can to make my students understand. True teachers are underappreciated. Much love.
This is the most crisp, lucid, and downright amazing explanation of the concept with demonstration. I could make out the efforts you have taken in designing this video, answering every detail of the concept and covering all possible scenarios every step of the way. You're hands down brilliant, thank you for this creation!!
Luis, you are a freakin’ genius 💜 There’s no way someone would not be able to understand any math concept if taught in such a lucid way.
Sir, it’s a humble request from my side (and I’m sure a lot of others would agree with me) to create an ordered playlist on math for machine learning, it’d be of great help!
Thank you Debarchan! I have some playlists on the front page of the channel (ua-cam.com/users/LuisSerrano), on supervised/unsupervised learning, probability, etc. Also, I organized all the videos into topics here: serrano.academy. Let me know if they help, or if you have any different suggestions. Cheers!
@@SerranoAcademy Nice
Much agreed. He's a genius!
So comprehensive yet always to the point. Amazing quality. My professor mentioned in passing that the Mars rovers perform SVD and only select the first few largest singular values to send back and you explained perfectly why this works!
This is the definition of perfect! I can't imagine anyone can explain these concepts better than him! Really appreciate it!
I have been struggling with the concept of SVD for months and now I finally get it with this clear and friendly tutorial. Thank you!
Probably the best explanation of the SVD technique that I've seen so far. Good job!
Having to compose a mathematical essay about the usage of SVD in image compression, without having been taught about the concept of linear algebra, I had to look for appropriate sources that would help me understand the theory. After hours of online research I managed to find your video and I have to admit that I'm impressed. Your analysis is both thorough and comprehensive in the simplest way possible. Thank you so much for your help! Keep up the good work!
Thank you, Mr. Luis Serrano. Much appreciated: the entire effort to put out the video with so much dexterity...AWESOME!!
Thank you Viswapriya!
My professor taught this in class and I was like what is the use of this. After watching this, I realised how fascinating this stuff is. Thanks a lot for making this. Shared with my friends.
You have saved me man, I had a presentation tomorrow on SVD, and now I have got everything that I need to know.
This is hands down the best explanation of SVD on the internet. Thank you
I've spent 3 hours finding this amazingly clear explanation instead of studying messy course notes and it worthed it:) Thank you, Mr. Serrano.
i have been finding study material to learn SVD and in this journey I went through several PDFs, slides and few books but this video is the best and crisp explanation of this algorithm. Thanks!
You are one of the best teachers i have ever come across. Please keep helping people like me with awesome explanations. May god bless you !
Two Thumbs Up!!!! I have been all over UA-cam the past two days watching presentations on SVD (many of them very good), and I am only 3:45 into this video and I **NOW** understand why there are two rotations involved with this simple illustration. This was gold and very, very helpful. Thank you!👍👍
Okay, mind blown a second time here at minute 13:00. Seeing the visual for why when the second scaling factor is zero it compresses the line and thus reduces the dimension via the degenerate transformation. Again, I've seen the math elsewhere, but not until the visual here did I actually understand what was going on. Thank you again.
I'm so happy to finally find you....the best...it was like you were answering my questions while I'm asking you in a live...hoping for more similar videos ❤️
By far this is the most clear explanation about SVD ! Greatly appreciate your work and efforts
Best video in UA-cam about SVD. Need more ML related videos. Keep it up :)
This is the best, the best, the best explanation of SVD, image compression I could find on youtube. Thank you so much Mr. Serrano!
Wow, you explained SVD in such a nice and clear manner! Love it. Thank you so much.
I've watched several videos trying to explain what linear transformations are and this one hit home for me...thank you!
Sheer brilliance. Your clarity of thought is mind bending. Luis serrano, statquest, ritvickmath, krish naik and codebasics - this is pretty much my holy grail for datascience
By far the best explanation I have seen for SVD. It could not be better than this. Thanks a lot Luis for this excellent video.
Easily the best explanation on youtube by far - I've been going all over trying to understand what SVD actually means. This video provided that and so much more, will catch your other videos and hopefully they're as good!
This explanation is genius. After watching a few videos on the subject, I got to this one and wow, everything I just saw made sense. A few people have this talent, and it’s awesome that you’re helping others with it. Keep up with your content!
This is extraordinary! Thanks a lot for simplifying the visual intuition behind SVD so clearly.
Usually for each topic that I search for, I watch couple videos, but hey dear youtuber just land here, he did the best explanation, thanks
The first half of the video is one of the best intruductions on SVD that I could find so far. Thank you so much.
this is the best explanation on SVD ive seen thank you! you really explain the logic behind it well
This IS the best explanation of ANYTHING (no matter what field, what topic) I've ever got in my life about something. Incredibly visual didactic style. Grandioso!
You did a phenomenal job on this concept and video! This is the best explanation I have seen as to why we can drop smaller singular values and end up with a reasonable approximation!
You're an excellent teacher, Luis. Can't thank you enough for the amazing explanations for these hard-to-grasp concepts. Hope you and your family are doing well during this pandemic.
Thank you! Hope you and your family are well too!
With due respect , but what Dr Boyd and Dr Strang could not help me understand, you made it so simple to get in for a lifetime... Thank you so much..
This is literally amazing. Please keep your work.
Wow! Seriusly this is perfect! You go direct to the root of it. Thank you a lot for making it easy to understand 😀😀
Wow! This is the most clear explanation on SVD.
This is truly an awesome tutorial. Thanks Luis!
I particularly love your explanation for image compression...just amazing! Thanks much!
Thank You Luis Serrano for making such concepts very easy to understand.
So informative and easy to follow. I love this. Thank you so much for taking the time to create this video. It's so important to know how the concepts we learn in class can be applied in real life. This has changed everything for me. Thank you again.
El mejor video de la vida para entendelo todo! SVD y PCA los aprendi gracias a los videos de este canal. Muchas Gracias.
Thank you sir for simplifying the depth of topic with such visuals under 30 mins 🙏. Now I have practical intuition of SVD
Wow.... You took a very complex topic and turned it easy-peasy. Thank you.
Amazing! All so plain and simple to see and understand the full concept of ML.
I never understood SVD until now. You are my favorite teacher :D
Absolutely brilliant. My brain absorbed it like a sponge. Will buy the book. Looks just fine for application without assuming any background
Very well explained! please upload some practical videos including coding as well.
World needs more people like you.
This is the best presentation on SVD ever seen.
Great video, it's really nice to see that kind of explanation out there :) Keep it up
BRAVO! If anyone of you viewers wonder what "disruption" looks like when it comes to technology and business models, this video is a perfect example. If we just narrow it down to getting the knowledge of, say SVD, then why on earth would anyone pay a tuition fee, commute for hours or relocate, to sit in front of a mediocre "teacher" who do not have the interest or pedagogical skills to teach (we have all had these "teachers" and it sucks! Is that not why you are here?!?!), when you can watch a video of this quality with prefect explanations and examples at the comfort of your home as many times as you want for free(*)??? Eventually no one will!!! Education is truly democratized one video at a time...
This video deserves more recognition!
You are most amazing person that know how to teach to make it intuitive.
I really love the way you explain these Maths topics which I never understand in Linear Algebra classes as theory but here I got in this one video with clear visualization and moreover we can always follow in python code which is really cool to see. Thanks for effort and time for strangers like me. Just Keep going.
Wow .it's so crazy and the mathematics representation was expressed in the pictorial representation of the beginning of this video is so awesome...keep rocking
Luis, gracias por compartir tus videos. Realmente son de lo mejor de YT. Siempre recomiendo que te sigan(te descubrí en Udacity)...me encanta la forma en que abordas SVD, es la misma forma que uso yo mismo: abordarla desde un punto de vista geométrico. Evidentemente, la calidad del video me deja sin palabras...la magia de SVD es la reducción de dimensionalidad y la muestras magistralmente, la clave de esta reducción la dices en 11:33 determinante = 0...¡Bravo!
This video is the most intuitive way of explaining the svd, thanks!
The best explanation & depiction of SVD.
You are Democratizing quality technical education and making it accessible to all.
Thank you! I came across SVD while doing natural language processing. This is the best explanation I have seen by far. Is there any chance you can do a chapter on SVD in the context of machine learning? Thank you again for the excellent work
One of the best explanation for SVD .. To good.
this was an awesome explanation. please do more of these . thanks a lot. please do one on Independent Component Analysis (ICA)
Wow... Much much better than any book or professor's explanation..
Mr. Luis Serrano you are the BEST.
Awesome awesome awesome illustration. Thank you, you made the idea very clear and simple
One of the finest explanation of singular value decomposition. Thank you!
Thank you so much this. I am really happy I found your channel.
I have been struggling to understand SVD.
Just found your channel. I need to understand SVD for my work and this is great.
One of the best and simple explanation of svd, thank u!
Your video made help us made our entire presentation for college - Thanks a tonne!
I guess this is pretty important concept! Gonna watch it ❤️ thanks for sharing such awesome content!
Thank you Rushiraj!
Best Explanation for intuition underlying SVDs
That's a great explanatory video. Thank you for making it crystal clear.
What is your Patreon? This channel...if you keep going like this is much better than 3B1B. Perfectly done! That, Luis is a freaking awesome channel. Your’s is much better!! Do Variance-Covariance. Thanks.
This explanation was absolutely amazing. Thank you so much for your help!
Wow! A very much simple explaination of SVD with great intution and visualization! The best explaination of SVD I have found in internet so far. Thank you sir, also, could you please make a video about the computation of SVD in a detailed manner.
This reminds me of residual vector quantization. Respect, Luis Serrano!
Thank you ! You´re explaining this concept in such a perfect way with so many good examples and visiuals. Keep up the good work :)
The best visual explanation of SVD.
Thank you so much, i have been struggling to understand SVD and finally I understood it well.
Thank you so much for making SVD simpler, loved it,
Indeed amazing demonstration of complex things!
Great video. Suggestion: You discussed square matrix in details in the first 20 minutes, which looks more like eigendecomposition. SVD needs to be discussed truly for rectangular matrix for another 30 minutes. Your graphic presentation is the best I saw in youtube on this topic so far.
Yeah, you could go the projection route: get mean, center dataset, get x and y variance, calculate the covariance matrix. Get eigenvector/eigenvalue set. Choose the bigger eigenvectors that correspond with larger eigenvaluees. Project on lower dimension.
I am thinking from which point should I begin writing the notes because everything in the entire video seems so fascinating and resourceful that I cannot skip anything.
The best explanation for SVD.Please do a video explaining latent semantic analysis and non-negative matrix factorisation
Thank you Kavya! I have one on matrix factorization here: ua-cam.com/video/ZspR5PZemcs/v-deo.html
Haven't done latent semantic analysis yet, but there's one on latent dirichlet allocation: ua-cam.com/video/T05t-SqKArY/v-deo.html
Dude you are soo good at explaining things. Keep it up!!!!!!!!!!!!!!!
For Dr. Serrano : Vox Luis, Vox Dei 🙏
Whenever I lack inspiration, I re-VIEW these gem videos!
Thank you for such a kind comment, what an honor. :) I'm glad you're enjoying the videos!
By far the best SVD explanation!
This video was amazing! Best explanation of the SVD I've seen so far.
After searching for this topic, I had a list of videos to choose from, then I remembered the great explanations I had gotten from this channel, so I selected this video... wasn't disappointed.
One of the best explanations ever!
Best explanation on the Internet
The best explanation of svd ever !
Hands down, the best video ever for svd. Thank you so much, Sir.
¡Gracias!
Gracias de nuevo por tu generosidad @PedroTrujilloV !
Great explaining such critical concept SVD ,
Always clever in making complex concepts , easy and simple
Thank you Serrano
Thank you Mohamed!
Pretty cool! I forgot a lot of my math studies in the last year, but your explenations were so precisely, but also simple that I understood nearly everything. Thank you (and sorry for my bad english :-) )
great video on SVD, helped me understand the concept really well, thank you very much 🙏